Resilient animals are minimally affected by disturbances, such as diseases and heat stress, and quickly recover. Daily activity data can potentially indicate resilience, because resilient animals likely keep variations due to disturbances that threat animal homeostasis at a low magnitude. We used daily step count of cows to define resilience indicators based on theory, exploratory analysis and literature, and then investigated if they can be used to genetically improve resilience by estimating heritability and repeatability, and genetic associations with other resilience-related traits, i.e. health traits, longevity, fertility, and body condition score (BCS). Two groups of resilience indicators were defined: indicators describing (1) mean step count at different lactation stages for individual cows, and (2) fluctuations in step count from individual step count curves. Heritability estimates were highest for resilience indicators describing mean step count, from 0.22 for the 2-week period pre-partum to 0.45 for the whole lactation. High mean step count was consistently, but weakly, genetically correlated with good health, fertility, and longevity, and high BCS. Heritability estimates of resilience indicators describing fluctuations ranged from 0.01 for number of step count drops to 0.15 for the mean of negative residuals from individual curves. Genetic correlations with health traits, longevity, fertility, and BCS were mostly weak, but were moderate and favorable for autocorrelation of residuals (− 0.33 to − 0.44) and number of step count drops (− 0.44 to − 0.56) with hoof health, fertility, and BCS. Resilience indicators describing variability of residuals and mean of negative residuals showed strong genetic correlations with mean step count (0.86 to 0.95, absolute), which suggests that adjustment for step count level is needed. After adjustment, ‘mean of negative residuals’ was highly genetically correlated with hoof health, fertility, and BCS. Mean step count, autocorrelation and mean of negative residuals showed most potential as resilience indicators based on resilience theory, heritability, and genetic associations with health, fertility, and body condition score. Other resilience indicators were heritable, but had unfavorable genetic correlations with several health traits. This study is an important first step in the exploration of the use of activity data to breed more resilient livestock.

中文翻译：

---

基于每日步数数据的奶牛育种复原力指标性状开发

复原力强的动物受疾病和热应激等干扰的影响最小，并能迅速恢复。日常活动数据可能表明复原力，因为复原力动物可能会由于威胁动物稳态的干扰而保持变化。我们使用奶牛的每日步数来定义基于理论、探索性分析和文献的复原力指标，然后通过估计遗传力和可重复性以及与其他复原力相关性状（即健康）的遗传关联来调查它们是否可用于从基因上提高复原力性状、寿命、生育能力和身体状况评分（BCS）。定义了两组恢复力指标：描述（1）个体奶牛不同泌乳阶段平均步数的指标，(2) 各个步数曲线的步数波动。描述平均步数的弹性指标的遗传力估计值最高，从产前 2 周的 0.22 到整个哺乳期的 0.45。高平均步数与良好的健康、生育能力和长寿以及高 BCS 具有一致但微弱的遗传相关性。描述波动的弹性指标的遗传力估计范围从步数下降的 0.01 到单个曲线的负残差平均值的 0.15。与健康特征、寿命、生育力和 BCS 的遗传相关性大多较弱，但对残差 (- 0.33 至 - 0.44) 和步数下降次数 (- 0.44 至 - 0.56) 与蹄健康、生育力的自相关是中等且有利的，和 BCS。描述残差变异性和负残差平均值的弹性指标显示出与平均步数（0.86 到 0.95，绝对值）的强遗传相关性，这表明需要对步数水平进行调整。调整后，“负残差的平均值”与蹄健康、生育能力和 BCS 高度相关。平均步数、自相关和负残差的平均值显示了作为弹性指标的最大潜力，这些指标基于弹性理论、遗传力以及与健康、生育能力和身体状况评分的遗传关联。其他复原力指标是可遗传的，但与一些健康特征具有不利的遗传相关性。这项研究是探索使用活动数据培育更具弹性的牲畜的重要的第一步。