Department of Agriculture, Food and Nutritional Science, University of Alberta, Edmonton, Alberta, Canada, T6G 2P5.
J Anim Sci. 2010 Dec;88(12):3789-99. doi: 10.2527/jas.2009-2407. Epub 2010 Sep 3.
This study sought to assess whether residual feed intake (RFI) calculated by regressing feed intake (DMI) on growth rate (ADG) and metabolic mid-BW in 3 different ways led to similar estimates of genetic parameters and variance components for young growing cattle tested for feed intake in fall and winter seasons. A total of 378 beef steers in 5 cohorts were fed a typical high energy feedlot diet and had free-choice access to feed and water. Feed intake data were collected in fall or winter seasons. Climate data were obtained from the University of Alberta Kinsella meteorological station and Vikings AGCM station. Individual animal RFI was obtained by either fitting a regression model to each test group separately (RFI(C)), fitting a regression model to pooled data consisting of all cohorts but including test group as a fixed effect (RFI(O)), or fitting a regression to pooled data with test group as a fixed effect but within seasonal (fall-winter or winter-spring) groups (RFI(S)). Two animal models (M1 and M2) that differed by the inclusion of fixed effects of test group or season, respectively, were used to evaluate RFI measurements. Feed intake was correlated with air temperature, relative humidity, solar radiation, and wind speed (-0.26, 0.23, 0.30, -0.14 for fall-winter and 0.31, -0.04, 0.14, 0.16 for winter-spring, respectively), but the nature and magnitude of the correlations were different for the 2 seasons. Single trait direct heritability, model likelihood, direct genetic variance, and EBV accuracy estimates were greatest for RFI(C) and least for RFI(O) for both M1 and M2 models. A significant genetic correlation was also observed between RFI(O) and ADG, but not for RFI(C) and RFI(S). Including a season effect (M2) in the genetic evaluation of RFI(O) resulted in the smallest heritability, model LogL, EBV accuracy, and largest residual variance estimates. These results, though not conclusive, suggest a possible effect of seasonality on feed intake and thus feed efficiency.
本研究旨在评估通过回归法分别估算生长速度（ADG）和代谢中间体重（BW）来计算残余采食量（RFI），是否会对秋冬季节进行采食量测试的年轻生长牛的遗传参数和方差分量产生相似的估计。共有 5 组 378 头肉牛，饲喂典型的高能饲养日粮，可自由选择采食和饮水。在秋冬季节采集采食量数据。气候数据来自阿尔伯塔大学金斯纳气象站和 Vikings AGCM 站。通过分别为每个测试组拟合回归模型（RFI(C)）、为包含所有组但包括测试组作为固定效应的汇总数据拟合回归模型（RFI(O)），或为包含测试组作为固定效应但在季节性（秋冬或冬春）组内拟合回归模型（RFI(S)），获得个体动物的 RFI。使用两种动物模型（M1 和 M2），分别通过包含测试组或季节的固定效应来评估 RFI 测量值。采食量与空气温度、相对湿度、太阳辐射和风速呈负相关（秋冬分别为-0.26、0.23、0.30、-0.14，冬春分别为 0.31、-0.04、0.14、0.16），但这两个季节的相关性质和程度不同。对于 M1 和 M2 模型，RFI(C)的单一性状直接遗传力、模型似然值、直接遗传方差和 EBV 准确性估计值最大，而 RFI(O)的则最小。RFI(O)和 ADG 之间也存在显著的遗传相关性，但 RFI(C)和 RFI(S)之间不存在。在 RFI(O)的遗传评估中包含季节效应（M2），会导致最小的遗传力、模型 LogL、EBV 准确性和最大的残差方差估计值。尽管这些结果还没有定论，但它们表明季节性对采食量和因此对饲料效率可能有影响。
