Clayton David
Cambridge Institute for Medical Research, University of Cambridge, Wellcome Trust/MRC Building, Addenbrookes's Hospital, Cambridge, UK.
Biostatistics. 2008 Oct;9(4):593-600. doi: 10.1093/biostatistics/kxn007. Epub 2008 Apr 25.
The problem of testing for genotype-phenotype association with loci on the X chromosome in mixed-sex samples has received surprisingly little attention. A simple test can be constructed by counting alleles, with males contributing a single allele and females 2. This approach does assume not only Hardy-Weinberg equilibrium in the population from which the study subjects are sampled but also, perhaps, an unrealistic alternative hypothesis. This paper proposes 1 and 2 degree-of-freedom tests for association which do not assume Hardy-Weinberg equilibrium and which treat males as homozygous females. The proposed method remains valid when phenotype varies between sexes, provided the allele frequency does not, and avoids the loss of power resulting from stratification by sex in such circumstances.
在混合性别的样本中检测X染色体上基因座的基因型与表型关联的问题,令人惊讶地很少受到关注。通过对等位基因进行计数可以构建一个简单的检验方法,男性贡献一个等位基因,女性贡献两个等位基因。这种方法不仅假设从研究对象所抽取的人群处于哈迪-温伯格平衡状态,而且可能还假设了一个不切实际的备择假设。本文提出了用于关联分析的1自由度和2自由度检验,这些检验不假设哈迪-温伯格平衡,并且将男性视为纯合子女性。当表型在不同性别之间存在差异时,只要等位基因频率不变,所提出的方法仍然有效,并且避免了在这种情况下因按性别分层而导致的检验效能损失。