【摘要】 目的 利用基因组结构方程模型(genome structural equation modeling,GSEM )研究阿尔茨海默病(Alzheimer’s disease,AD)和8种心血管疾病(cardiovascular disease,CVD)之间的共享遗传结构。方法 本研究基于大型联盟和全基因组关联研究中的AD和CVD数据，运用连锁不平衡分数回归评估疾病之间的遗传相关性;引入GSEM，在奇数染色体进行探索性因子分析、偶数染色体进行验证性因子分析来交叉验证，并结合路径分析建立三因子模型、共因子模型和双因素模型，从而揭示各疾病之间复杂的遗传关系。此外，还将奇偶染色体互换位置进行了相互验证。结果 所选取的9种疾病可以被3个独立的因子所构成的三因子模型所解释;共因子模型可解释一个共因子与各疾病之间的关系;扩展的双因素模型(三因子模型和共因子模型的结合)在拟合研究数据时表现最佳。结论 GSEM可用于探索AD与CVD之间的共享遗传结构以发现两者的相互作用，为理解复杂疾病之间的遗传关系奠定了方法学基础。更多还原

【Abstract】 Objective Genome structural equation modeling(GSEM) was used to study the shared genetic structure between Alzheimer’s disease(AD) and eight cardiovascular diseases(CVD).Methods In this study,based on AD and CVD data from large consortia and genome-wide association studies,we used chained unbalanced score regression to assess the genetic correlation between the diseases;GSEM was introduced to cross-validate with exploratory factor analysis at odd chromosomes and validation factor analysis at even chromosomes,and pathway analysis was combined to establish three-factor,cofactor,and bifactorial models,so as to reveal the complex genetic relationships between the diseases.Additionally,a cross-validation was performed by swapping the positions of odd and even chromosomes.Results The nine selected diseases can be explained by a three-factor model consisting of three independent factors;a cofactor model explains the relationship between a co factor and each disease;and an extended two factor model(a combination of the three-factor model and the cofactor model) performs best in fitting the study data.Conclusion GSEM can be used to explore the shared genetic structure between AD and CVD to discover their interactions,laying a methodological foundation for understanding the genetic relationship between complex diseases.更多还原