Miller Mark P, Haig Susan M, Ballou Jonathan D, Steel E Ashley
US Geological Survey Forest and Rangeland Ecosystem Science Center, 3200 SW Jefferson Way, Corvallis, OR 97331; Smithsonian National Zoological Park, Conservation Biology Institute, Front Royal, VA; and USDA Forest Service Pacific Northwest Research Station, Seattle, WA.
J Hered. 2017 Jul 1;108(5):574-582. doi: 10.1093/jhered/esx032.
Understanding and estimating inbreeding is essential for managing threatened and endangered wildlife populations. However, determination of inbreeding rates in natural populations is confounded by incomplete parentage information. We present an approach for quantifying inbreeding rates for populations with incomplete parentage information. The approach exploits knowledge of pedigree configurations that lead to inbreeding coefficients of F = 0.25 and F = 0.125, allowing for quantification of Pr(I|k): the probability of observing pedigree I given the fraction of known parents (k). We developed analytical expressions under simplifying assumptions that define properties and behavior of inbreeding rate estimators for varying values of k. We demonstrated that inbreeding is overestimated if Pr(I|k) is not taken into consideration and that bias is primarily influenced by k. By contrast, our new estimator, incorporating Pr(I|k), is unbiased over a wide range of values of k that may be observed in empirical studies. Stochastic computer simulations that allowed complex inter- and intragenerational inbreeding produced similar results. We illustrate the effects that accounting for Pr(I|k) can have in empirical data by revisiting published analyses of Arabian oryx (Oryx leucoryx) and Red deer (Cervus elaphus). Our results demonstrate that incomplete pedigrees are not barriers for quantifying inbreeding in wild populations. Application of our approach will permit a better understanding of the role that inbreeding plays in the dynamics of populations of threatened and endangered species and may help refine our understanding of inbreeding avoidance mechanisms in the wild.
了解和估计近亲繁殖对于管理受威胁和濒危的野生动物种群至关重要。然而，自然种群中近亲繁殖率的确定因亲本信息不完整而变得复杂。我们提出了一种方法，用于量化亲本信息不完整的种群的近亲繁殖率。该方法利用了导致近亲繁殖系数F = 0.25和F = 0.125的谱系配置知识，从而能够量化Pr(I|k)：给定已知亲本比例(k)时观察到谱系I的概率。我们在简化假设下推导了分析表达式，这些表达式定义了不同k值下近亲繁殖率估计器的属性和行为。我们证明，如果不考虑Pr(I|k)，近亲繁殖会被高估，并且偏差主要受k的影响。相比之下，我们纳入Pr(I|k)的新估计器在实证研究中可能观察到的广泛k值范围内是无偏的。允许复杂的代际间和代内近亲繁殖的随机计算机模拟产生了类似的结果。我们通过重新审视已发表的阿拉伯羚羊（Oryx leucoryx）和马鹿（Cervus elaphus）的分析，说明了考虑Pr(I|k)对实证数据可能产生的影响。我们的结果表明，不完整的谱系并不是量化野生种群近亲繁殖的障碍。应用我们的方法将有助于更好地理解近亲繁殖在受威胁和濒危物种种群动态中所起的作用，并可能有助于完善我们对野生环境中近亲繁殖避免机制的理解。
