مرتضی مختاری

The aim was to estimate the relative contribution of imprinting effects from both paternal and maternal sides to phenotypic variation in milk production traits including 305days milk yield (MY), average daily milk production (ADM), fat percentage (F%), protein percentage (P%), 305days fat yield (FY), 305days pro tein yield (PY), ratio of fat percentage to protein percentage (F:P) and somatic cell score (SCS) in Iranian Holstein cows. To do this, each trait was analysed with a series of four animal models, which were identical for fixed and additive genetic effects but differed for combinations of paternal and maternal imprint ing effects. The log-likelihood ratio test (LRT) and Akaike's information criteria (AIC) were used to select the best model for each trait. Correlations between traits due to additive and imprinting effects were estimated by bivariate analyses. For all traits studied, fitting the imprinting effect led to a better data fit. Also, it resulted in a noticeable decrease in additive genetic variance from 8% (SCS) to 28% (F:P). A significant maternal imprinting effect was detected on all traits studied. Estimates of maternal imprinting heritability (h2mi) were 0.07±0.02, 0.04±0.01, 0.06±0.01, 0.05±0.01, 0.5±0.01, 0.09±0.02, 0.07±0.02 and 0.06±0.01 for MY, ADM, F%, P%, FY, PY, F:P and SCS, respectively. For F:P, in addition to the ma ternal imprinting effect, a significant paternal imprinting component was also detected with a 7% contribution to phenotypic variance of F:P. Estimates of direct heritability (h2a) were 0.29±0.02, 0.17±0.01, 0.22±0.02, 0.11±0.01, 0.18±0.02, 0.22±0.02, 0.15±0.04 and 0.06±0.01 for MY, ADM, F%, P%, FY, PY, F:P and SCS, respectively. Maternal imprinting correlations (rmi) were in a wide range between −0.75±0.15 (P%-SCS) and 0.95±0.11 (MY-ADM). Additive genetic correlations (ra) ranged between −0.54±0.05 (MY-P%) and 0.99±0.01 (MY-ADM) and phenotypic correlations (rp) ranged from −0.30±0.01 (MY-F%) to 0.93±0.01 (MY ADM). The Spearman's correlation between additive breeding values including and excluding imprinting effects deviated from unity especially for top-ranked animals implying re-ranking of top animals following the inclusion of imprinting effects in the model. Since including imprinting effects in the model resulted in better data fit and re-ranking of top animals, including these effects in the genetic evaluation models for milk production traits was recommended