Zhang Feng, Deng Hong-Wen
College of Medicine, Xi'an Jiaotong University, 710061 Xi'an, People's Republic of China.
Genetica. 2010 Oct;138(9-10):945-50. doi: 10.1007/s10709-010-9476-6. Epub 2010 Aug 1.
Cryptic relatedness was suggested to be an important source of confounding in population-based association studies (PBAS). The impact of cryptic relatedness on the performance of haplotype phase inference and haplotype-based association tests is not clear. In this study, we used the Hapmap genetic data to simulate a set of related samples. We evaluated the accuracy of haplotype phase inferred by PHASE 2.1 and calculated the power, type I error rates, accuracy and positive prediction value (PPV) of haplotype frequency-based association tests (HFAT) and haplotype similarity-based association tests (HSAT) under various scenarios, considering relatedness levels, disease models and sample sizes. Cryptic relatedness appeared to slightly increase the accuracy of haplotype phase inference. We observed significant negative effect of cryptic relatedness on the performance of HFAT and HSAT. Ignoring cryptic relatedness may increase spurious association results in haplotype-based PBAS.
隐秘相关性被认为是基于人群的关联研究（PBAS）中一个重要的混杂因素来源。隐秘相关性对单倍型相位推断和基于单倍型的关联检验性能的影响尚不清楚。在本研究中，我们使用Hapmap遗传数据模拟了一组相关样本。我们评估了PHASE 2.1推断的单倍型相位的准确性，并计算了基于单倍型频率的关联检验（HFAT）和基于单倍型相似性的关联检验（HSAT）在各种情况下的检验效能、I型错误率、准确性和阳性预测值（PPV），同时考虑了相关性水平、疾病模型和样本量。隐秘相关性似乎略微提高了单倍型相位推断的准确性。我们观察到隐秘相关性对HFAT和HSAT的性能有显著的负面影响。在基于单倍型的PBAS中，忽略隐秘相关性可能会增加虚假关联结果。
