Estimation of (co)variance components and genetic parameters for litters per sow per year and pigs weaned per sow per year

Increasing the number of pigs weaned per sow per year is an economically important goal for pig producers. This study was conducted to estimate (co)variance components and genetic parameters of pigs weaned per sow per year (PWSY) and litters per sow per year (LSY), and then determine the possibility of direct selection for improving both traits. Data analyzed were from 11,283 sows from three herd groups, which included 1,925 sows from a purebred herd in Thailand, and 9,358 sows from two purebred/crossbred herd groups in the United States over a time period ending in March 2007. The distribution of both traits was considered to be normally distributed, based on the shape of the normal curve of each trait in each herd group. The (co)variance components, heritabilities, and genetic correlations for each trait were estimated using Restricted Maximum Likelihood (REML) bivariate animal model analysis. The animal model for LSY and PWSY contained fixed effects for breed, parity, and year-month of farrowing (contemporary group). Estimates of heritability (h 2) for LSY were low and quite similar, with the values between 0.09 and 0.18 across the three herd groups. The heritability estimates (h 2) for PWSY were low to moderate around 20 percent across the three herd groups. The estimated genetic correlation of LSY with PWSY was high and favorable, ranging from 0.45 to 0.70 across purebred and crossbred herds. The opportunity to make permanent genetic improvement by including LSY and PWSY in the selection program appears to be possible because the heritability of LSY and PWSY was low to moderate, and genetic correlation between two traits was favorable. Therefore, selection for genetic improvement in LSY or PWSY appears possible and should complement each other.