论文部分内容阅读
目的提出一种基于置换检验(permutation test)和主成分分析(PCA)的检验方法-permutation-PCA。探讨在病例对照关联分析中permutation-PCA在不同遗传模型下的表现。方法基于“首吸飘感”对照组数据分别模拟产生7种不同遗传模型下的病例对照基因分型数据,采用permutation-PCA方法对模拟数据进行检验,并进一步用permuta-tion-PCA方法检验“首吸飘感”基因分型实际数据。结果Permutation-PCA方法对7种不同遗传模型假设下的模拟病例对照基因分型数据的检验结果差异均具有统计学意义(P均<0.05)。Permutation-PCA方法对于“首吸飘感”基因分型实际数据的检验结果差异具有统计学意义(P<0.05)。结论Permutation-PCA方法对不同遗传模型假设不敏感,适用于复杂疾病关联分析中病例对照基因分型数据的研究。
Objective To propose a test method based on permutation test and principal component analysis (PCA) -permutation-PCA. To investigate the performance of permutation-PCA in different genetic models in case-control association analysis. Methods The case-control genotyping data of 7 different genetic models were simulated based on the data of the first-in-first-breath-control group. The data were tested by permutation-PCA method and further analyzed by permuta-tion-PCA method Test “first suction floating sense ” Genotyping actual data. Results The results of Permutation-PCA test showed that there were significant differences in the test results of simulated case-control genotypes under the assumption of seven different genetic models (all P <0.05). The Permutation-PCA method has significant statistical difference (P <0.05) in the test results of the actual data of the “first-invaded influenza” genotyping. Conclusion The Permutation-PCA method is insensitive to the hypothesis of different genetic models and is suitable for the study of case-control genotyping data in the association analysis of complex diseases.