摘要
Feed costs are reported as the largest variable expense in beef production systems accounting for 50% to 70% of total production costs. Due to the large impact of feed costs on profitability, producers have become increasingly aware of the need to improve feed utilization in cattle. With the advent of technologies to measure individual feed intake in cattle, a phenotype is available for selection; however, how to implement this phenotype into a breeding objective or genetic evaluation is debatable. This dissertation examines some of these unanswered questions about feed intake and efficiency and its utility in for genetic selection. The objectives herein were 1) to evaluate the maternal genetic effects on dry matter feed intake (DMI), 2) to simulate data to examine the effects single trait selection on DMI or residual feed intake (RFI) on genetically correlated traits of weaning weight (WWT) and yearling weight (YWT), 3) to examine data generated by an ear tag accelerometer (CowManger; Agis Automatisering BV, Harmelen, Netherlands) while attached to steers located in a feedlot and on pastures to develop a proxy for measures of intake and 4) to examine the phenotypic relationship between grazing and feedlot intake.
For the examination of maternal genetic effects on intake, the American Gelbvieh Association (AGA) and the Red Angus Association of America (RAAA) provided pedigree information in addition to DMI and WWT records. Dry matter intake records were limited to animals within an age range of 240 d to 365 d to limit data to only postweaning cattle. Embryo transfer calves were removed. Gelbvieh and Red Angus data were analyzed separately.
The first analysis was a single trait model that examined the maternal genetic effects of DMI. Contemporary groups (CG) were formed using sex, pen, feed trial designation, trial length and year for both AGA and RAAA. The final data set for AGA consisted of 3,021 animals with DMI records and a 3-generation pedigree of 15,418 animals. For Red Angus, cattle with DMI records was 3,213 and a 3-generation pedigree of 13,747 animals. The heritabilities of DMI direct for Gelbvieh and Red Angus were moderate to high (0.45 0.06 and 0.24 0.06, respectively) but the DMI maternal heritability for Gelbvieh was 0.00 0.00 and Red Angus DMI maternal heritability was very low at 0.05 0.04. Resulting in little to no maternal effect for DMI.
The second analysis was a multi-trait model estimating the correlation between DMI and WWT maternal. For the multi-trait model, fixed effects for weaning weight included age, age of dam and CG. The contemporary group for WWT was sex, breeder, weaning date, and herd. For Gelbvieh, the heritabilities for DMI, WWT direct and WWT maternal were 0.45 0.05, 0.36 0.06 and 0.15 0.05, respectively. The heritabilities for Red Angus were 0.27 0.05, 0.21 0.06 and 0.16 .07 for DMI, WWT direct and WWT maternal, respectively. The genetic correlation between DMI and WWT maternal was low at 0.12 0.13 and 0.12 0.24, for Gelbvieh and Red Angus, respectively. These results suggested that WWT maternal would has minimal impact on the estimation of DMI EPD using a multivariate model.
For the second objective of this dissertation, data were simulated to examine the effects on genetically correlated traits of WWT and YWT when single trait selection was conducted on DMI or RFI. Genetic parameters were established using published variance estimates weighted for the number of animals included in each estimation. Based on the weighted genetic parameters, a simulated population was established for three selection scenarios. The first and second scenario for selection was single trait selection for DMI and RFI, respectively. The third scenario used an economic selection index as criteria for replacements. With an annual replacement rate of 20% for females and 5% for males, 10 years of offspring data was generated. Replacements were chosen based on their breeding value for the trait of interest. At the conclusion of 10 years of simulated data, the scenarios for the selection of DMI and RFI saw a decrease in DMI of 0.85 kg/year and 0.19 kg/year, respectively. Both scenarios also resulted in a decrease for YWT of 27.83 kg/year from selection of DMI and 5.13 kg/year from the selection of RFI. Selection using the economic index showed a steady increase in YWT (14.84 kg/year) but also demonstrated an increase of DMI (0.42 kg/year). The three scenarios were all examined by the amount of profitability determined from fed cattle and feed prices. Of the three scenarios, the economic index showed the greatest amount of profit due to the increase in YWT. Although DMI increased with the index, the amount of increase in yearling weights was significant enough to outweigh the increase in feed costs.
The third study of this dissertation examined a remote sensor technology as a potential proxy for DMI in addition to estimating a correlation between DMI measured in a feedlot versus grazing intake for cattle on pasture. Ninety-three steers were equipped with a CowManager ear tag accelerometer (CME) that measured the amount of time an animal spent ruminating, eating, and levels of activity. The steers were placed in a feedlot where their intakes were measured using the GrowSafe Feed Intake monitoring system. The data collected via CME and GrowSafe were analyzed to identify existing associations between the measurements. Based on the DMI measured by GrowSafe, the 15 highest and 15 lowest intake animals were identified. These low/high intake animals (LHI) were used to quantify grazing intake using the biomarker titanium dioxide (TiO2). No association between CME and DMI measured in the feedlot were found. Pearson’s correlations for CME measurements and DMI were low and ranged from -0.11 to 0.12. A regression analysis found no significant CME variable as explanatory variables for DMI.
After a 54-d performance test, the steers were immediately transported to pasture where the steers were maintained for 43 days. Data from CME were continuously collected while cattle grazed on pasture. For the final 20 d, the LHI cattle were administered a bolus of 10 g of TiO2 each morning. During the 6 final days of this study, rectal fecal samples were collected twice daily with collection occurring 12 h apart. Every 24 h, time of collection was advanced 2 hours to minimize effects of diurnal variation. The fecal samples were analyzed for TiO2 concentration and based on these concentrations grazing dry matter intakes (GDMI) were estimated. The Pearson’s correlation between GDMI and DMI measured in the feedlot was 0.84 ± 0.10 (P < 0.05) with a Spearman rank correlation of 0.99 ± 0.03 (P < 0.05). This result suggested a strong relationship between grazing and feedlot DMI; however, it is less than 1 indicating some change in rank between DMI and GDMI. The correlation between GDMI and CME ranged from -0.22 to 0.19 with the largest correlation (-0.22) was between GDMI and time spent eating. All of these correlation estimates were not significantly different from zero (P > 0.05). This study was able to show the application of remote sensor technology for monitoring cattle maintained on rangeland, but the precision of measurements from CME failed to provide an indicator for GDMI. A strong relationship between feedlot measured DMI and DMI for grazing cattle was established in this study.
摘要译文
据报道，饲料成本是牛肉生产系统中最大的可变费用，占总生产成本的50％至70％。由于饲料成本对获利能力的巨大影响，生产者越来越意识到提高牛饲料利用率的必要性。随着测量牛个人饲料摄入量的技术的出现，有一个表型可供选择；然而，如何将这种表型应用于育种目标或遗传评估尚待商.。本文研究了一些关于饲料采食量和效率及其在遗传选择中的效用的悬而未决的问题。本文的目的是：1）评估母体对干物质采食量（DMI）的遗传效应，2）模拟数据，以检验单性状选择对DMI或剩余采食量（RFI）对断奶体重遗传相关性状的影响（ WWT）和一岁体重（YWT），3）检查耳标加速度计（CowManger；荷兰Harmelen的Agis Automatisering BV公司）附着在饲养场和牧场上的ste牛皮上，以制定摄食量的替代指标； 4）检查放牧和育肥场摄食量之间的表型关系。 ，美国Gelbvieh协会（AGA）和美国红安格斯协会（RAAA）除了提供DMI和WWT记录外，还提供了谱系信息。干物质摄入记录仅限于240 d至365 d年龄范围内的动物，以将数据限制为仅断奶后的牛。去除胚胎移植小牛。分别分析了Gelbvieh和Red Angus的数据。第一个分析是检验DMI的母体遗传效应的单性状模型。 AGA和RAAA的性别，围栏，饲料试验名称，试验时间和年份组成了当代人群（CG）。 AGA的最终数据集包括3021只具有DMI记录的动物和15418只动物的3代谱系。对于红安格斯来说，有DMI记录的牛是3,213，而3代谱系中有13,747只动物。直接Gelbvieh和红安格斯的DMI遗传性中等至高（分别为0.450.06和0.240.06），但是Delmi的DMI母体遗传性为0.00 her 0.00，而Red Angus DMI的母体遗传性很低，为0.050.04。导致DMI的产妇影响很小甚至没有。第二项分析是估计DMI与WWT产妇之间相关性的多特征模型。对于多特征模型，断奶体重的固定影响包括年龄，水坝年龄和CG。 WWT的当代群体是性别，繁殖者，断奶日期和畜群。对于盖尔维耶夫（Gelbvieh），DMI，直接WWT和WWT母体的遗传力分别为0.45，0.05、0.360.06和0.150.05。对于DMI，WWT直接和WWT母体，红安格斯的遗传力分别为0.27±0.05、0.21±0.06和0.16±0.07。吉尔比维（Gelbvieh）和红安格斯（Red Angus）的DMI和WWT孕妇的遗传相关性较低，分别为0.120.13和0.120.24。这些结果表明，WWT产妇对采用多变量模型估算DMI EPD的影响最小。针对本研究的第二个目标，模拟数据以检验进行单性状选择时对WWT和YWT遗传相关性状的影响在DMI或RFI上。使用公开的方差估计值建立遗传参数，加权的方差估计值包含在每个估计值中的动物数量。基于加权遗传参数，针对三个选择方案建立了模拟种群。选择的第一种和第二种情况分别是DMI和RFI的单性状选择。第三种情况使用经济选择指数作为替代标准。雌性和雌性的年替代率分别为20％和5％，生成了10年的后代数据。根据替代品的育种价值来选择感兴趣的替代品。在10年模拟数据的总结中，选择DMI和RFI的方案使DMI分别降低了0.85千克/年和0.19千克/年。两种情况下，DMI的选择也使YWT减少了27.83公斤/年，RFI的选择减少了5.13公斤/年。使用经济指数进行选择显示出YWT稳定增加（14.84千克/年），但DMI也有所增加（0.42千克/年）。三种情况均通过根据饲喂牛和饲料价格确定的获利能力进行了检验。在这三种情况下，经济指数显示出最多的利润，这是由于YWT的增加。尽管DMI随指数增加，但一岁体重的增加量足以抵消饲料成本的增加。本论文的第三项研究考察了遥感技术作为DMI的潜在替代品，除了估计了DMI之间的相关性外。测量于