Zintzaras Elias, Santos Mauro
Department of Biomathematics, University of Thessaly School of Medicine, 2 Panepistimiou Str, Larissa 41110, Greece.
Stat Appl Genet Mol Biol. 2012 Jun 27;11(4):Article 4. doi: 10.1515/1544-6115.1804.
We evaluate power performance to detect the correct mode of inheritance in gene-disease associations of two different approaches: the MAX test and the degree of dominance index or h-index. The MAX test is a special case of the conditional independence tests that simultaneously test for association and select the most likely genetic model based on a three-dimensional normal distribution. The h-index is based on the philosophy of using orthogonal contrasts to infer the mode of inheritance quantitatively. A population genetic model is developed where the real mode of inheritance is known a priori and power performance can be accurately determined. The simulations showed that none of the two approaches generally outperforms the other, nor each of them provides a panacea to estimate efficiently the mode of inheritance in all parameter space. However, the simultaneous application of both approaches can provide insights in determining the underlying mode of inheritance.
我们评估了两种不同方法（MAX检验和显性指数或h指数）在基因-疾病关联中检测正确遗传模式的效能表现。MAX检验是条件独立性检验的一种特殊情况，它基于三维正态分布同时检验关联性并选择最可能的遗传模型。h指数基于使用正交对比定量推断遗传模式的理念。我们构建了一个群体遗传模型，其中先验已知真实的遗传模式，从而可以准确确定效能表现。模拟结果表明，这两种方法一般都不会普遍优于另一种方法，也没有一种方法能在所有参数空间中提供有效估计遗传模式的万灵药。然而，同时应用这两种方法可以为确定潜在的遗传模式提供见解。
