Longitudinal genetic dissection of feed efficiency and feeding behaviour in MaxGro pigs

The aim of this study was to examine the genetic characteristics of feed efficiency and feeding behaviour traits longitudinally along with their associations. Data were available on 3027 purebred MaxGro pigs from 52 (11, SD) to 110 (11, SD) Kg BW in 7 weeks on test. Longitudinal phenotypes of average daily feed intake (FI), occupation time in automatic feeder (OT), number of visit (NV) and feeding rate (FR) were obtained per each week on test. Feed intake and behaviour records were obtained by electronic feeder (IVOG) under ad libitum feeding condition and all traits were expressed as weekly averages. To obtain genetic association of feed efficiency and feeding behaviour traits, residual feed intake (RFI) was estimated as FI adjusted phenotypically for ADG of the entire test period and lean meat percentage (LMP) at end of the test. Data were analysed using random regression model containing animal genetic effect, permanent environmental effect of each pig and contemporary group using Legendre polynomials of weeks on test along with heterogeneous residual variances. Furthermore, the covariance function was used to estimate the genetic parameters of above traits at different weeks on test. Posterior mean of heritability estimates of FI and RFI showed quadratic changes during test period with lowest in week one (0.15 (standard deviation, 0.03) and 0.07 (0.02), respectively) to highest in week three (0.31 (0.03) and 0.24 (0.04), respectively). Posterior mean of heritability estimates of weekly OT (0.32 (0.04) to 0.37 (0.04)), NV (0.17 (0.03) to 0.22 (0.03)) and FR (0.26 (0.05) to 0.31 (0.04)) were consistent at different stages of growth. Posterior mean of genetic correlations among RFI estimates dropped by increase in weeks on test to −0.06 (0.16) between first and last weeks on test. However, positive large genetic correlations were observed among OT 0.79–0.99), NV 0.77–0.99) and FR 0.87–0.99). Furthermore, RFI showed varying genetic correlations with feeding behaviour traits of OT, NV and FR with early stages having (0.17, −0.23 and 0.38, respectively) and late stages showing (0.72, 0.25 and −0.07, respectively). Modelling the dynamics of feed efficiency may enhance the genetic improvement of feed efficiency due to considering the change in feed intake capacity of pigs. Inclusion of feeding behaviour traits to pig breeding programs may further improve the genetic progress of feed efficiency due to their larger heritability estimates than RFI and their existing correlations with feed efficiency.