Crosby Ralph W, Williams Tiffani L
Department of Computer Science, College of Charleston, Charleston, SC, USA.
Department of Computer Science, Northeastern University, Charlotte, NC, USA.
BMC Bioinformatics. 2017 Dec 6;18(Suppl 15):514. doi: 10.1186/s12859-017-1916-1.
The inference of species divergence time is a key step in most phylogenetic studies. Methods have been available for the last ten years to perform the inference, but the performance of the methods does not yet scale well to studies with hundreds of taxa and thousands of DNA base pairs. For example a study of 349 primate taxa was estimated to require over 9 months of processing time. In this work, we present a new algorithm, AncestralAge, that significantly improves the performance of the divergence time process.
As part of AncestralAge, we demonstrate a new method for the computation of phylogenetic likelihood and our experiments show a 90% improvement in likelihood computation time on the aforementioned dataset of 349 primates taxa with over 60,000 DNA base pairs. Additionally, we show that our new method for the computation of the Bayesian prior on node ages reduces the running time for this computation on the 349 taxa dataset by 99%.
Through the use of these new algorithms we open up the ability to perform divergence time inference on large phylogenetic studies.
物种分化时间的推断是大多数系统发育研究中的关键步骤。在过去十年中已有进行此类推断的方法，但这些方法的性能对于包含数百个分类单元和数千个DNA碱基对的研究而言，扩展效果仍不尽人意。例如，一项针对349个灵长类分类单元的研究估计需要超过9个月的处理时间。在这项工作中，我们提出了一种新算法AncestralAge，它显著提高了分化时间推断过程的性能。
作为AncestralAge的一部分，我们展示了一种计算系统发育似然性的新方法，我们的实验表明，在上述包含超过60,000个DNA碱基对的349个灵长类分类单元数据集上，似然性计算时间提高了90%。此外，我们表明我们用于计算节点年龄贝叶斯先验的新方法在349个分类单元数据集上的此计算运行时间减少了99%。
通过使用这些新算法，我们开启了在大型系统发育研究中进行分化时间推断的能力。
Zotero 插件
