Inferring the Roadmap of Bacterial Gene Transfer

描述（由申请人提供）：横向基因转移是对细菌基因组进化几乎无处不在的影响，引入了广泛的新遗传物质，可以促进适应；在病原体中尤其如此，基因组岛引入了代谢功能，允许采用致病的生活方式。虽然大量的横向转移基因在受体基因组中被识别，但关于潜在供体池的信息有限。由于横向基因转移是如此强大的改变力量，供体分类群的性质——包括它们的系统发育关系和代谢能力——为理解现存谱系的进化提供了一个背景，也为深入了解它们的进化潜力提供了一个背景。重要的是，有几个理论和经验的论点，为什么供体分类群可能不是随机从所有细菌中选择的，而是可能与受体更密切相关。这样，细菌分支可能是在基因转移的影响下形成的，而不是简单地反映了共享基因的共同祖先。我们建议为最近获得的主要细菌分支的基因库确定可能的供体分类群。首先，将使用人工基因组框架开发一种鲁棒方法，用于利用和组合参数方法进行外来基因检测。其次，我们将开发基因聚类和基因组分割方法来识别外来基因，使用人工基因组和上述参数指南。这些方法将识别出彼此相似的非典型基因组。这个想法是具有相似特性的非典型基因可能起源于同一来源。利用这些基因簇，我们将通过外推系统发育方法来推断数据库中没有同源基因的供体分类群，从而确定它们可能的祖先。供体分类群的分配将通过一种新的复制偏倚模式分析来验证，使我们能够将外来基因的链偏倚模式与其供体进化枝的基因组特征联系起来，这在系统发育上是相当广泛的。这个框架将允许我们检验基因转移更经常发生在更密切相关的谱系之间的假设。

DESCRIPTION (provided by applicant): Lateral gene transfer is a nearly ubiquitous influence on bacterial genome evolution, introducing a wide palette of novel genetic material the can facilitate adaptation; this is especially true in pathogens, where genomic islands have introduced metabolic functions allowing the adoption of the pathogenic lifestyle. While a large number of laterally transferred genes are recognized in recipient genomes, information regarding potential donor pools is limited. Since lateral gene transfer is such a powerful force for change, the nature of donor taxa - including their phylogenetic relationships and metabolic capabilities - provides a context for understanding the evolution of extant lineages, as well as insight into their evolutionary potential. Critically, there are several theoretical and empirical arguments for why donor taxa may not be chosen at random from all bacteria, but rather may be more closely related to the recipients. In this way, bacterial clades may form under the influence of gene transfer, and may not simply reflect the common ancestry of shared genes. We propose to identify the likely donor taxa for the recently-acquired gene pools of major bacterial clades. First, an artificial genome framework will be used to develop a robust method for utilizing and combining parametric methods for alien gene detection. Second, we will develop both gene cluster and genome segmentation methods for the identification of alien genes, using artificial genomes and parametric guidelines established above. These methods will identify groups of atypical genes that are similar to one another. The idea is that atypical genes with similar properties may have originated from the same source. Using these clusters of genes, we will ascertain their likely ancestry by extrapolating phylogenetic methods to infer the donor taxa for genes without homologues in the database. The assignment of donor taxa will be verified by a novel analysis of replication bias patterns, allowing us to correlate strand-bias patterns of alien genes with the signatures of the genomes in their donor's clades, which can be phylogenetically quite broad. This framework will allow us to test the hypothesis that gene transfer occurs more often between more closely related lineages.