Bayesian model adequacy and choice in phylogenetics
Bayesian model adequacy and choice in phylogenetics
复制标题
DOI：
10.1093/oxfordjournals.molbev.a004175
复制
发表时间：
2002-07-01
影响因子：
10.7
通讯作者：
Bollback, JP
中科院分区：
文献类型：
--
作者：
Bollback, JP
贝叶斯推理正在成为一种常见的系统发育估计的统计方法，原因之一是它允许对具有复杂进化模型的大型数据集进行快速分析。方便的是，贝叶斯系统发育方法使用了目前可用的序列进化随机模型。然而，与其他基于模型的方法一样，贝叶斯推理的结果取决于假设的进化模型：不充分的模型（与数据不太吻合的模型）可能导致错误的推断。在本文中，我提出了一种贝叶斯系统发育方法，该方法使用后验预测分布来评估进化模型的充分性。通过评估模型的后验预测性能，可以选择合适的模型进行贝叶斯系统发育研究。尽管我提供了一个评估系统发育模型的整体（全局）性能的测试统计量，但是可以定制各种测试统计量来评估进化模型的特定特征（局部性能），以识别故障的来源。与似然比检验和参数自举不同，本文提出的方法考虑了系统发育和模型参数的不确定性。
Bayesian inference is becoming a common statistical approach to phylogenetic estimation because, among other reasons, it allows for rapid analysis of large data sets with complex evolutionary models. Conveniently, Bayesian phylogenetic methods use currently available stochastic models of sequence evolution. However, as with other model-based approaches, the results of Bayesian inference are conditional on the assumed model of evolution: inadequate models (models that poorly fit the data) may result in erroneous inferences. In this article, I present a Bayesian phylogenetic method that evaluates the adequacy of evolutionary models using posterior predictive distributions. By evaluating a model's posterior predictive performance, an adequate model can be selected for a Bayesian phylogenetic study. Although I present a single test statistic that assesses the overall (global) performance of a phylogenetic model, a variety of test statistics can be tailored to evaluate specific features (local performance) of evolutionary models to identify sources failure. The method presented here, unlike the likelihood-ratio test and parametric bootstrap, accounts for uncertainty in the phylogeny and model parameters.