McArdle John J, Prescott Carol A
Department of Psychology, University of Southern California, Los Angeles, CA
Department of Psychology, University of Southern California, Los Angeles, CA.
Perspect Psychol Sci. 2010 Sep;5(5):606-21. doi: 10.1177/1745691610383510.
There is a great deal of interest in the analysis of Genotype × Environment interactions (G×E). There are some limitations in the typical models for the analysis of G×E, including well-known statistical problems in identifying interactions and unobserved heterogeneity of persons across groups. The impact of a treatment may depend on the level of an unobserved variable, and this variation may dampen the estimated impact of treatment. Some researchers have noted that genetic variation may sometimes account for unobserved, and hence unaccounted for, heterogeneity. The statistical power associated with the G×E design has been studied in many different ways, and most results show that the small effects expected require relatively large or nonrepresentative samples (i.e., extreme groups). In this article, we describe some alternative approaches, such as randomized designs with multiple measures, multiple groups, multiple occasions, and analyses, to identify latent (unobserved) classes of people. These approaches are illustrated with data from the Aging, Demographics, and Memory Study (part of the Health and Retirement Study) examining the relations among episodic memory (based on word recall), APOE4 genotype, and educational attainment (as a proxy for an environmental exposure). Randomized clinical trials (RCTs) and randomized field trials (RFTs) have multiple strengths in the estimation of causal influences, and we discuss how measured genotypes can be incorporated into these designs. Use of these contemporary modeling techniques often requires different kinds of data be collected and encourages the formation of parsimonious models with fewer overall parameters, allowing specific G×E hypotheses to be investigated with a reasonable statistical foundation.
人们对基因型×环境交互作用（G×E）的分析有着浓厚的兴趣。G×E分析的典型模型存在一些局限性，包括在识别交互作用时存在的众所周知的统计问题以及各组人群中未观察到的异质性。一种治疗方法的效果可能取决于一个未观察到的变量的水平，而这种变化可能会削弱对治疗效果的估计。一些研究人员指出，基因变异有时可能解释未观察到的、因而未被考虑到的异质性。与G×E设计相关的统计功效已经通过多种不同方式进行了研究，大多数结果表明，预期的小效应需要相对较大或不具代表性的样本（即极端组）。在本文中，我们描述了一些替代方法，如具有多种测量、多组、多次场合和分析的随机设计，以识别潜在的（未观察到的）人群类别。通过来自老龄化、人口统计学和记忆研究（健康与退休研究的一部分）的数据说明了这些方法，该研究考察了情景记忆（基于单词回忆）、APOE4基因型和教育程度（作为环境暴露的替代指标）之间的关系。随机临床试验（RCT）和随机现场试验（RFT）在因果影响估计方面有多种优势，我们讨论了如何将测量的基因型纳入这些设计中。使用这些当代建模技术通常需要收集不同类型的数据，并鼓励形成总体参数较少的简约模型，从而能够在合理的统计基础上研究特定的G×E假设。
