Genetic analysis of clinical mastitis in different risk periods by linear and threshold models
Avainsanat:
mastitis, dairy cattle, bayesian analysis, genetic parametersAbstrakti
Understanding the genetic basis of clinical mastitis (CM) during lactation is essential in deciding on the best measure of CM for breeding value evaluation. CM is one of the leading reasons for premature disposal of dairy cows. Culled cows have incomplete lactation records, and in statistical analysis this may cause sampling biases, because culled cows are seldom a random sample. It is, therefore, essential to study the effects of culling information on the genetic analysis of CM. The objectives of this study were first: to estimate heritability and genetic association between CM during the different stages of first lactation using linear and threshold models, second: to assess the effects culling information on the estimates of variance (co)variance components and associated genetic parameters.
First lactation was divided into six lactation stages (risk periods) by days (d) after calving: CM1, (-7 to 150 d); CM2, (-30 to 30 d); CM3, (-30 to 150 d); CM4, (31 to 150 d); CM5, (150 to 300 d); CM6, (-30 to 300 d). To assess the effect of including or excluding records of cows that had been culled from the herd before the end of the risk period, two data sets were prepared. In the first data set (Data I), records of cows culled before the end of the risk period were not included. In the second data set (Data II), records of cows culled during the risk period were included; if the cow had been culled for mastitis reasons or had completed two-thirds of the opportunity period if culled for other reasons. Variance (co)variance components were estimated by linear and threshold-liability models employing Bayesian approach.
Heritability estimates using the linear model ranged from 0.005 to 0.024 for Data I, and from 0.005 to 0.029 for Data II, depending on the stages of lactation or risk period defined. The corresponding estimates from the threshold-liability model ranged from 0.034 to 0.076, and from 0.043 to 0.083 for Data I and Data II, respectively. In general, increased sire variance and higher heritability were observed with the inclusion of culling information. Estimated genetic correlations between the CM traits were medium to high, and ranged from 0.42 to 0.99 in Data I, and from 0.61 to 0.99 in Data II. Thus CM cannot be regarded the same trait in the course of lactation.
Results from the threshold analyses showed that the heritability of liability to CM was highest for CM2 and followed by CM6, CM3 and CM1with estimates of 0.083, 0.073, 0.073 and 0.072, respectively. Estimates for CM4 and CM 5 were the lowest with heritability of 0.043 and 0.047, respectively. The high heritability estimate for CM2 indicates that most genetic information is in early lactation. Thus, the best measure of CM would be to consider only cases in early lactation. However, as the genetic correlation between CM traits defined over the different risk periods were less than one, mastitis cannot be regarded as the same trait in the different parts of lactation. Thus, a multivariate model, treating mastitis in different stages of lactation as different traits, would be the best model for sire evaluation.
