Lange Christoph, DeMeo Dawn L, Laird Nan M
Department of Biostatistics, Harvard School of Public Health, Boston, MA 02115, USA.
Am J Hum Genet. 2002 Dec;71(6):1330-41. doi: 10.1086/344696. Epub 2002 Nov 21.
In the present article, we address family-based association tests (FBATs) for quantitative traits. We propose an approach to analytical power and sample-size calculations for general FBATs; this approach can be applied to virtually any scenario (missing parental information, multiple offspring per family, etc.). The power calculations are used to discuss optimal choices of the phenotypes for the FBAT statistic and its power's dependence on ascertainment conditions, on study design, and on the correct specification of the distributional assumptions for the phenotypes. We also compare the general FBAT approach with PDT and QTDT. The practical relevance of our theoretical considerations is illustrated by their application to an asthma study.
在本文中,我们探讨了用于定量性状的基于家系的关联检验(FBATs)。我们提出了一种用于一般FBATs的分析效能和样本量计算的方法;该方法几乎可应用于任何情况(父母信息缺失、每个家庭有多个后代等)。效能计算用于讨论FBAT统计量的表型的最佳选择,以及其效能对确定条件、研究设计和表型分布假设的正确设定的依赖性。我们还将一般FBAT方法与PDT和QTDT进行了比较。通过将我们的理论考量应用于一项哮喘研究,说明了其实际相关性。