Motivation: The risk of many complex diseases is determined by an interplay of genetic and environmental factors. The examination of gene-environment interactions (G×Es) for multiple traits can yield valuable insights about the etiology of the disease and increase power in detecting disease-associated genes. However, the methods for testing G×Es for multiple traits are very limited. Method: We developed novel approaches to test G×Es for multiple traits in sequencing association studies. We first perform a transformation of multiple traits by using either principal component analysis or standardization analysis. Then, we detect the effects of G×Es using novel proposed tests: testing the effect of an optimally weighted combination of G×Es (TOW-GE) and/or variable weight TOW-GE (VW-TOW-GE). Finally, we employ Fisher’s combination test to combine the p values. Results: Extensive simulation studies show that the type I error rates of the proposed methods are well controlled. Compared to the interaction sequence kernel association test (ISKAT), TOW-GE is more powerful when there are only rare risk and protective variants; VW-TOW-GE is more powerful when there are both rare and common variants. Both TOW-GE and VW-TOW-GE are robust to directions of effects of causal G×Es. Application to the COPDGene Study demonstrates that our proposed methods are very effective. Conclusions: Our proposed methods are useful tools in the identification of G×Es for multiple traits. The proposed methods can be used not only to identify G×Es for common variants, but also for rare variants. Therefore, they can be employed in identifying G×Es in both genome-wide association studies and next-generation sequencing data analyses.
Hum Hered

中文翻译：

---

在测序关联研究中测试多种性状的基因-环境相互作用。


动机：许多复杂疾病的风险是由遗传和环境因素的相互作用决定的。检查多种性状的基因-环境相互作用 (G×E) 可以产生有关疾病病因的宝贵见解，并提高检测疾病相关基因的能力。然而，测试多个性状的 G×E 的方法非常有限。方法：我们开发了新方法来测试测序关联研究中多个性状的 G×E。我们首先通过使用主成分分析或标准化分析来执行多个性状的转换。然后，我们使用新提出的测试来检测 G×E 的效果：测试 G×E 的最佳加权组合 (TOW-GE) 和/或可变权重 TOW-GE (VW-TOW-GE) 的效果。最后，我们采用 Fisher 组合检验来组合p值。结果：大量的仿真研究表明所提出的方法的 I 类错误率得到了很好的控制。与交互序列核关联测试（ISKAT）相比，当只有罕见的风险和保护性变异时，TOW-GE 更强大；当同时存在罕见和常见变体时，VW-TOW-GE 的功能更强大。 TOW-GE 和 VW-TOW-GE 对于因果 G×E 的影响方向都是稳健的。 COPDGene 研究的应用表明我们提出的方法非常有效。结论：我们提出的方法是识别多个性状的 G×E 的有用工具。所提出的方法不仅可以用于识别常见变异的 G×E，还可以用于识别罕见变异。 因此，它们可用于在全基因组关联研究和下一代测序数据分析中识别 G×E。
 赫里德