结构方程混合模型在SNP分析中的应用

采用结构方程混合模型(SEMM)对实际SNP数据进行分析,为遗传统计学提供一种新的有效的分析方法。本研究的数据是由GAW17提供的,包含697个个体的22条常染色体的上万个SNP和根据这些SNP所模拟的697个个体的性状特点。随机挑选了1号染色体上的4个SNP和3个定量性状作为研究变量,分别进行潜在类别分析和结构方程混合模型分析。根据4个SNP数据,人群被分为3个潜在类别,概率分别为0.53,0.34,0.13。潜在类别1、2和3中的因子均值Q分别为-4.029、-2.052和0,潜在类别1、2的因子均值均低于3(<0.001)。研究表明:结构方程混合模型(SEMM)综合了结构方程模型和潜在类别模型的思想,形成了自己的优势,可用于处理同时包含分类潜变量和连续潜变量的数据。

Structural equation mixture modeling and their application in analysis of SNP

To analyze SNP data of GAW17 by Structural equation mixture modeling （SEMM）, and to provide a new method for the study of genetic statistic . The data is provided by GAW17, it contains 697 individual, 22 autonom- ic tens of thousands of SNP and the SNP simulated 697 individual trait characteristics. In this study, randomly se- lected the four SNP from chromosome 1 and three quantitative traits as a research variable, which were analysised by latent class and mixed structural equation modeling. According to four SNP data, the crowd was divided into three potential categories, each category probability were O. 53,0.34, 0.13. Factors mean Q of latent class 1,2 and 3 are -4.029, -2.052 and 0. We knew that factor mean of latent class 1, 2 are lower than 3 （ 〈0. 001 ）. So we have reasons to think that structural equation mixed modeling integrated the structural equation modeling and latent class modeling thoughts, formed its own advantage, which can be used for processing classification latent variable and continuous latent variable data.