Cardiovascular disease risk
Cardiovascular disease risk
复制标题
DOI：
10.1038/nm0496-365
复制
发表时间：
1996-04
期刊：
影响因子：
82.9
通讯作者：
David Curtis
中科院分区：
文献类型：
--
作者：
David Curtis
致编辑：描述冠状动脉疾病（CAD）与内皮型一氧化氮合酶基因（ecNOS）多态性之间关系的关联研究在统计学上存在缺陷，因此从结果中得出的任何结论都应该比作者声称的更不确定。Wang等人忽略了这样一个事实，即在所有的分子遗传研究中都可能存在大量的基因座，在关联研究中，需要测试多态性。这个问题最近在连锁研究的背景下进行了讨论，尽管我批评这些指南过于保守（DC，提交的手稿），但要点仍然是，当使用传统的显著性水平时，许多结果将被证明是假阳性。关联研究的问题可能更加严重，因为只测试了一个很小的遗传区域，而多个独立的多态性可能存在于同一区域。本研究的作者报告说，他们是第四个被检测与CAD相关的ecNOS多态性。它
To the editor-The association study'describing a relationship between coronary artery disease (CAD) and a polymorphism of the endothelial nitric oxide synthase gene (ecNOS) has statistical weaknesses such that any conclusions drawn from the results should be more tentative than those claimed by the authors. Wang et aL have ignored the fact that in all molecular genetic investigations there are potentially a huge number of loci and, in the case of association studies, polymorphisms to be tested. This issue has recently been discussed in the context of linkage studies', and although I have criticized these guidelines as being overly conservative (DC, manuscript submitted), the point remains that when conventional levels of significance are used, many results will turn out to be false-positives. With association studies the problem may be even more severe because only a tiny genetic region is tested, and multiple independent polymorphisms may exist in the same region. The authors of the present study report that theirs was the fourth ecNOS polymorphism to be tested for association with CAD. It