非正态验证性因子分析在基因整体效应中的应用  

Non-normal confirmatory factor analysis in the application of the whole gene effect

作　　者：刘小琴[1] 马瑞[1] 罗艳虹[1] 李治[1] 张春森[1] 张岩波[1] 

机构地区：[1]山西医科大学公共卫生学院卫生学统计教研室,山西太原030001

出　　处：《生物信息学》2013年第3期192-195,共4页Chinese Journal of Bioinformatics

基　　金：国家自然科学基金资助项目(31071156)

摘　　要：针对SNPs数据不服从正态分布的情况,拟采用S-B测度调整估计方法拟合验证性因子模型,进行SNPs整体效应和关联性分析。用GAW17提供的SNPs数据进行实例分析。本研究随机选取2号染色体上,分布在6个基因之中的13个SNPs作为研究对象,对选取的6个基因做潜变量得分,然后对基因和疾病感染做检验。结果显示:χ2/df最大似然估计方法的卡方自由度比为3.59,S-B测度调整估计方法的卡方自由度比χ2/df为2.89,最大似然估计方法的RMSEA为0.061,S-B测度调整估计方法的RMSEA为0.052。6个基因对该感染都有影响.由此得出结论,在处理SNPs数据时,使用S-B测度调整估计能得到更好的拟合模型。可以推测这6个基因下的13个SNP位点可能是感染的致病位点。This paper proposed S-B measure （scaled） estimates to fit confirmatory factor models, to analysis overall effect and correlation of SNPs which does not fit normal distribution. Example of SNPs data is provided by GAW17. The study chooses 13 SNPs located 6 gene in chromosome 2, we firstly do latent variables score in the six genes , and genes and infections do t-test. Maximum likelihood estimation Chi square degrees of freedom X^2/df is 3.59, S- B scaled method X^2/df/s 2.89, maximum likelihood estimation RMSEA is 0.061, S-B method RMSEA is 0. 052. Six genes on the infection have influence. When analysis the SNPs data , using S-B estimated can get a better fitted model. We can speculate that the 13 SNPs sites in 6 genes may be the infection pathogenic site.

关 键 词：单核苷酸多态性 非正态 最大似然估计 S-B测度调整估计 验证性因子分析 

分 类 号：Q75[生物学—分子生物学]