Genetic parameters for large-scale behavior traits and type traits in Charolais beef cows1

Abstract

In the last decades, beef cattle breeding mainly focused on improving production and reproduction traits. Nowadays, there is a growing interest to include behavior and type traits in the breeding goal. There is an interest in behavior as it is associated with human safety and workability and in type traits as they might be associated with longevity of cows. The objective of the current study was to estimate the heritability for behavior and type traits in Charolais and to estimate the genetic correlations among these traits. Behavior traits, including aggressiveness at parturition, aggressiveness during gestation period, and maternal care, were scored by farmers using an on-farm recording system to enable large-scale collection of phenotypes. Type traits, including udder traits ( = 3), teat traits (3), feet and leg traits (5), and locomotion (1), were scored by 10 trained classifiers. Data was available on 6,649 cows in parity 1 to 12 and located in 380 herds. Results showed that differences between herds explained up to 23% of the total phenotypic variance in behavior traits. This might be due to differences in management or to consistent differences in scoring between farmers. Aggressiveness at parturition had higher heritability (0.19) and higher genetic coefficient of variation (CV = 11%) than aggressiveness during gestation ( = 0.06 and CV = 4%) and maternal care ( = 0.02 and CV = 2%). Heritabilities for udder traits (0.14 to 0.20) and teat traits (0.17 to 0.35) were higher than for feet and leg traits (0.02 to 0.19). Genetic coefficients of variation for udder and teat traits were also higher (up to 21%) than for feet and leg traits (up to 11%). Strong genetic correlations were found between behavior traits (with absolute values from 0.71 to 0.98). The genetic correlations indicate that it is difficulty to simultaneously improve maternal care and reduce aggressiveness. We concluded that there are good opportunities to implement selection for improved udder and teat traits and against aggressiveness at parturition using a simple on-farm recording system of behavior.

Heritability and evolvability of fitness and nonfitness traits: Lessons from livestock

“…Although there is a wealth of information on h 2 and V A in livestock, with few notable exceptions (e.g., Sartori et al 2015;Vallee et al 2015), estimates of I A or other measures of evolvability are scarce. Thus h 2 remains the preferred dimensionless population parameter used in animal breeding (Hill and Kirkpatrick 2010) which along with genetic correlations with other traits are used to predict the extent to which a trait is altered by selection.…”

“…; Vallee et al. ). Livestock data are potentially valuable because animals are typically raised in rather homogenous production environments, genetic parameters (particularly for cattle data) are often estimated from thousands or even hundreds of thousands of individuals, and there are typically multiple studies from different researchers working in different countries providing consistent estimates that have been combined in reviews and metaanalyses (e.g., Bittante et al.…”

“…One way of dealing with imprecise estimates of genetic parameters and comparisons of data sets from unrelated organisms scored in different environments is to focus on situations where accurate estimates of heritability and evolvability are available from a species or a set of species measured in relatively similar environments. A particularly unique resource in this respect is provided by farmed livestock and particularly cattle, where there is a wealth of information on h 2 and V A but much less on I A or other measures of evolvability, with only a few exceptions (e.g., Sartori et al 2015;Vallee et al 2015). Livestock data are potentially valuable because animals are typically raised in rather homogenous production environments, genetic parameters (particularly for cattle data) are often estimated from thousands or even hundreds of thousands of individuals, and there are typically multiple studies from different researchers working in different countries providing consistent estimates that have been combined in reviews and metaanalyses (e.g., Bittante et al 2012;Berry and Crowley 2013;.…”

Selection for pure- and crossbred performance in Charolais

“…e ijkl is the random residual effect ~N(0,I e 2 ) where 2 e is the residual variance. In the GWAS analysis, variance components were fixed at values estimated from a larger data set described in Vallée et al (2015). Analyses were performed using ASReml ).…”

“…Locomotion was recorded on a scale from 1 representing lameness to 5 representing no lameness. More details on data collection are available in Vallée et al (2015). From the 3,274 cows scored for behavior and type traits, 3,064 cows were also scored for muscular development as a part of the national recording scheme.…”