Pooled parameter estimates for traits of economic importance in indigenous chicken in the tropics

Abstract

Genetic and phenotypic parameter estimates obtained in different studies, may be heterogeneous. This is because they are obtained in different locations, using different sample sizes and estimation methods. The choice of which parameter estimates to use, is in most cases subjective and this may lead to underestimation or overestimation of potential genetic gain. The objective of the current study was to use meta-analysis to estimate genetic and phenotypic parameters of traits of economic importance in indigenous chicken (IC) breeding program. Estimates from literature were included where it was clearly indicated that the study was conducted on indigenous chicken and in the tropics. Information captured included country of study, method of evaluation, sample size, mean and standard deviation or standard error of the estimate, and for genetic correlations, number of sires. Seventy three (73) heritabilities and 38 genetic and phenotypic correlations traits were used. A random effect model was used so as to include the variances within and between studies. Weighted heritabilities for body weight at hatch (Bw0) and twelve weeks (Bw12), egg number at twelve weeks (EN12) and antibody response (Ab) were 0.46, 0.24, 0.12, and 0.27 respectively. Pooled and non-pooled heritability values were varied for body weight among other traits for instance the pooled genetic value for body weight at hatch was 0.46 as compared to 0.56 in one of the studies. The weighted genetic and phenotypic correlations were also varied from the ones from single studies. Heritability estimates differed due to method of estimation, location and time at which the data was collected.