Oseni S, Tsuruta S, Misztal I, Rekaya R
Department of Animal and Dairy Science, The University of Georgia, Athens 30602-2771, USA.
J Dairy Sci. 2004 Dec;87(12):4327-33. doi: 10.3168/jds.S0022-0302(04)73578-X.
The influence of various editing criteria for days open (DO) records on genetic parameter estimates of DO and pregnancy rates (PR) in US Holsteins was investigated. Data included first parity 305-d milk yield and DO records from 8 states: Georgia (GA), Florida (FL), North Carolina (NC), Texas (TX), Arizona (AZ), California (CA), New York (NY), and Wisconsin (WI). The pregnancy rate was computed as 1/[(DO - VWP)/HI + 1)], where VWP was the approximate voluntary waiting period and HI was the heat interval set as 21 d. The upper limit for PR was set to 1.0. A bivariate animal model for DO (or PR) and 305-d milk yield was fit separately for each state. The model included fixed effects of herd-year, month of calving, and age of cow, as well as random animal and residual effects. In separate analyses, maximum DO records were limited to 150, 200, 250, 300, and 365 d. Analyses for PR used values of 50, 80, and 120 d for the VWP. Genetic and residual variances for DO were strongly dependent on the upper limit; both variances were 8 times larger as the upper bound increased from 150 to 365 d. Estimates of heritability for DO varied between 0.03 and 0.06. There was a 30% increase in the heritability estimate as the upper limit increased from 150 to 250 d for FL and NC, and small or no increases for the other states. The increase of the upper limit from 250 to 365 d resulted in little change. The genetic correlation between milk and DO was the highest for FL (0.6) and the lowest for GA (0.12 to 0.23). For PR with VWP=50, the heritability was higher than the corresponding estimate for DO in GA, equal to that in AZ, and lower in the remaining states. Heritabilities of PR also varied by the length of VWP; highest heritabilities were obtained at VWP=50 d for GA and AZ; at VWP=80 d for NY and WI; at VWP=120 d for FL, NC, and CA. Increase of genetic variation for records of DO < 250 d was small. Days open and PR are strongly influenced by differences in management protocols among states.
研究了美国荷斯坦奶牛不同的产犊间隔（DO）记录编辑标准对DO遗传参数估计值和妊娠率（PR）的影响。数据包括8个州头胎305天产奶量和DO记录：佐治亚州（GA）、佛罗里达州（FL）、北卡罗来纳州（NC）、得克萨斯州（TX）、亚利桑那州（AZ）、加利福尼亚州（CA）、纽约州（NY）和威斯康星州（WI）。妊娠率计算公式为1/[（DO - VWP）/HI + 1]，其中VWP为近似的自愿等待期，HI为设定为21天的发情间隔。PR的上限设定为1.0。针对每个州分别拟合了DO（或PR）与305天产奶量的二元动物模型。该模型包括牛群年份、产犊月份和奶牛年龄的固定效应，以及随机动物效应和残差效应。在单独的分析中，最大DO记录限制为150、200、250、300和365天。PR分析使用的VWP值为50、80和120天。DO的遗传方差和残差方差强烈依赖于上限；随着上限从150天增加到365天，这两个方差均增大8倍。DO的遗传力估计值在0.03至0.06之间变化。对于佛罗里达州和北卡罗来纳州，随着上限从150天增加到250天，遗传力估计值增加了30%，而其他州增加较小或没有增加。上限从250天增加到365天导致变化很小。产奶量与DO之间的遗传相关性在佛罗里达州最高（0.6），在佐治亚州最低（0.12至0.23）。对于VWP = 50时的PR，佐治亚州的遗传力高于相应的DO估计值，亚利桑那州与之相等, 其余州则较低。PR的遗传力也因VWP长度而异；佐治亚州和亚利桑那州在VWP = 50天时获得最高遗传力；纽约州和威斯康星州在VWP = 80天时获得最高遗传力；佛罗里达州、北卡罗来纳州和加利福尼亚州在VWP = 120天时获得最高遗传力。产犊间隔小于250天记录的遗传变异增加较小。产犊间隔和妊娠率受各州管理方案差异的强烈影响。
