Bayesian modeling to infer mechanisms of GWAS hypertension genes in rats

 描述（由申请人提供）：高血压是影响数百万美国人的心脏病和中风的主要危险因素。全基因组关联研究（GWAS）已经发现了大量与高血压相关的基因.这些基因中的许多在很大程度上是未知的，即使在细胞和动物模型系统中也使其影响的研究复杂化。我们提出了一种新的方法，这些基因的功能可以通过将它们置于致敏啮齿动物系统中的基因-基因相互作用的背景下来确定，Dahl盐敏感（SS）大鼠是一种成熟的啮齿动物高血压模型。我们将通过定义已知在盐敏感型高血压中具有功能重要性的单细胞类型的分子网络来降低复杂性，所述单细胞类型是肾外髓质的髓质粗升支（mTAL）的上皮Na+转运细胞。我们将SS衍生的原代培养mTAL细胞暴露于一组九种刺激物以诱导转录变化。暴露后定期采集样本，并测量其转录组。这些时间过程数据将使用一种新的贝叶斯图形方法进行分析，该方法将创建一个依赖于时间的基因-基因相互作用网络。该网络将用于预测SS基因组敲除的影响。这些预测将使用来自SS背景下四只GWAS提名敲除大鼠的原代培养mTAL进行测试。敲除对网络的影响将用于预测敲除大鼠的表型，这些表型将在这些动物模型中进行测试。通过我们的方法生成的网络将为GWAS提名基因提供背景，并将大大提高GWAS提名基因座转化为可测试和有用的动物模型，从而推进盐敏感性高血压遗传学的研究。

 DESCRIPTION (provided by applicant): Hypertension is a major risk factor for heart disease and stroke that affects millions of Americans. Genome- wide association studies (GWAS) have nominated a large number of genes as important in human hypertension. Many of these genes are largely unknown, complicating study of their effects even in cell and animal model systems. We propose a novel approach whereby the function of these genes may be determined by placing them into the context of gene-gene interactions in a sensitized rodent system, the Dahl salt- sensitive (SS) rat, a well-established rodent model of hypertension. We will reduce complexity by defining the molecular network of a single cell type known to be of functional importance in salt-sensitive forms of hypertension, the epithelial Na+ transporting cells of the medullary thick ascending limb (mTAL) of the renal outer medulla. We will expose SS-derived primary cultured mTAL cells to a panel of nine stimuli to induce transcriptional changes. Samples will be collected at regular intervals after exposure and their transcriptomes will be measured. These time-course data will be analyzed using a novel Bayesian graphical approach which will create a time-dependent gene-gene interaction network. This network will be used to make predictions about the effect of knockouts from the SS genome. These predications will be tested using primary cultured mTALs from four GWAS-nominated knockout rats on the SS background. The effects of the knockouts on the networks will be used to predict phenotypes for the knockout rats, which will be tested in these animal models. The network generated by our method will provide context for GWAS-nominated genes and will greatly improve the conversion of GWAS-nominated loci into testable and useful animal models, thereby advancing the study of the genetics of salt-sensitive hypertension.