• 主题
高级搜索  >
历史搜索 清空历史
首页> 美国卫生研究院文献>Evolutionary Bioinformatics Online >Using minimum bootstrap support for splits to construct confidence regions for trees
【2h】

Using minimum bootstrap support for splits to construct confidence regions for trees

机译：使用对分割的最小引导程序支持来构建树的置信区域

代理获取
本网站仅为用户提供外文OA文献查询和代理获取服务，本网站没有原文。下单后我们将采用程序或人工为您竭诚获取高质量的原文，但由于OA文献来源多样且变更频繁，仍可能出现获取不到、文献不完整或与标题不符等情况，如果获取不到我们将提供退款服务。请知悉。
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Many of the estimated topologies in phylogenetic studies are presented with the bootstrap support for each of the splits in the topology indicated. If phylogenetic estimation is unbiased, high bootstrap support for a split suggests that there is a good deal of certainty that the split actually is present in the tree and low bootstrap support suggests that one or more of the taxa on one side of the estimated split might in reality be located with taxa on the other side. In the latter case the follow-up questions about how many and which of the taxa could reasonably be incorrectly placed as well as where they might alternatively be placed are not addressed through the presented bootstrap support. We present here an algorithm that finds the set of all trees with minimum bootstrap support for their splits greater than some given value. The output is a ranked list of trees, ranked according to the minimum bootstrap supports for splits in the trees. The number of such trees and their topologies provides useful supplementary information in bootstrap analyses about the reasons for low bootstrap support for splits. We also present ways of quantifying low bootstrap support by considering the set of all topologies with minimum bootstrap greater than some quantity as providing a confidence region of topologies. Using a double bootstrap we are able to choose a cutoff so that the set of topologies with minimum bootstrap support for a split greater than that cutoff gives an approximate 95% confidence region. As with bootstrap support one advantage of the methods is that they are generally applicable to the wide variety of phylogenetic estimation methods.
机译：系统发育研究中许多估计的拓扑都对所示拓扑中的每个拆分提供了自举支持。如果系统发育估计无偏见，则对拆分的高引导程序支持表明有足够的确定性表明该拆分实际上存在于树中，而较低的引导程序支持则表明估计拆分的一侧可能有一个或多个类群实际上，另一端与分类单元位于同一位置。在后一种情况下，后续的问题是关于多少个分类单元以及哪些分类单元可能会被错误地放置以及它们可能放置在何处，这些问题并没有通过提供的引导程序来解决。我们在这里提出一种算法，该算法找到所有树木的集合，这些树木的最小自举支持大于大于给定值的拆分。输出是树的排序列表，根据树中拆分的最小引导程序支持排名。此类树的数量及其拓扑结构在引导程序分析中提供了有用的补充信息，有关引导程序对拆分的支持较低的原因。我们还提出了通过考虑所有具有最小引导程序大于某个数量的拓扑的集合来量化低引导程序支持的方法，以提供拓扑的置信区域。使用双引导程序，我们可以选择一个截止值，以便具有最小引导程序支持的，大于该截止值的分割的拓扑集给出大约95％的置信度区域。与自举支持一样，该方法的优点之一是它们通常可应用于各种各样的系统发育估计方法。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
代理获取

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号