三种SNPs数据仿真方法的效能比较  

Comparision of the efficacy of three SNPs data simulation methods

作　　者：刘芸良 肖纯[1] 史晓雯[1] 刘艳[1] 

机构地区：[1]哈尔滨医科大学卫生统计学教研室,黑龙江哈尔滨150081

出　　处：《实用预防医学》2018年第2期152-155,共4页Practical Preventive Medicine

基　　金：国家自然科学基金(30972537;81172741)

摘　　要：目的比较HAPGEN2、gs 2.0和GWAsimulator2三种方法在仿真单核苷酸多态性(SNPs)数据时的效能差异,为以后使用SNPs数据仿真方法提供指导。方法以真实人群SNPs数据作为原始数据,利用三种方法分别生成仿真数据,通过连锁不平衡模式和最小等位基因频率评价仿真效能,并通过χ~2差异位点评价致病位点的设置效能。结果 HAPGEN2仿真连锁不平衡模式的能力优于gs 2.0和GWAsimulator2,gs 2.0和GWAsimulator2仿真最小等位基因频率的能力近似且均优于HAPGEN2,三种方法均能良好的设置单致病位点。结论三种SNPs数据仿真方法均有优劣,用户可根据实际需求选择合适的仿真方法。Objective To compare the differences in the efficacy of HAPGEN2, gs2.0 and GWAsimulator2 in simulating single nucleotide polymorphisms （ SNPs } data, and to provide guidance for the future use of SNPs data simulation methods. Methods The simulation data were generated by the above-mentioned three methods using the real population SNPs data as the original data. The simulation performance was evaluated by the linkage disequilibrium model and the minimum allele frequency, and the setting efficiency of the disease site was evaluated by χ2 difference. Results The ability of HAPGEN2 to simulate the linkage disequilibrium model was superior to those of gs2.0 and GWAsimulator2. The abilities of gs2.0 and GWAsimulator2 to simulate the minimum allele frequency were similar to as well as superior to that of HAPGEN2, and all the three methods could be used to set the single disease site. Conclusions The three SNPs data simulation methods have advantages and disadvantages, and the us- ers can choose the appropriate simulation method according to the actual demand.

关 键 词：单核苷酸多态性 计算机仿真 连锁不平衡 最小等位基因频率 

分 类 号：R195.1[医药卫生—卫生统计学]