Social genetic effects on productive and feeding behavior traits in growing period of Duroc pigs

Aiming to explore the role of feeding behavior traits (FBt) and social genetic models for the genetic evaluation of both FBt and performance traits (Pt), genetic parameters were estimated for daily gain (DG), daily feed consumption (DC), feed conversion ratio (FCR), backfat thickness (BF), feeding rate (FR), feeding frequency (FF) and occupation time (OT). Traits were recorded in 663 Duroc growing pigs. Two types of bivariate models were fitted: animal models (AM) and social interaction animal models (SAM). Genetic parameters were estimated using Bayesian procedures. Heritability (h2) estimates obtained with AM for all traits were medium-high (0.25-0.6); higher estimates (0.40-1.27) of the ratio of total genetic variance to phenotypic variance (T2) were obtained from SAM. This is consequence of the additional heritable variation captured by social genetic effects (SGE). Only OT direct genetic effects (DGE) seem to be positively correlated with DGE of DG, DC and BF (0.34, 0.61 and 0.38, respectively), when AM was used the respective genetic correlations were not different from zero. Considering AM unfavorable genetic correlations between FCR and DG (0.42) and between BF and DG (0.64) were estimated. SAM evidenced that the correlation between FCR and DG seems to have its origin in the SGE. The correlation between DGE was null, and that for SGE was 0.51. With regard to those between BF and DG, either SGE or DGE correlations remained high and unfavorable, but the correlation between SGE of DG and DGE of BF was negative (-0.80), being null that between SGE of BF and DGE of DG. Large estimation errors of within-trait direct-social genetic correlations prevented to define their sign, but they seem to be of low magnitude. The role of FBt to improve Pt genetic evaluations is limited, except for OT. The consideration of SAM allows disentangling the social origin of certain unfavorable AM correlations. SAM could be used to explore indexes combining SGE and DGE of the different traits to take advantage of favorable genetic correlations that might exists between them.