基于蒙特卡洛模拟分析不同基因集方法的效能

Effect Analysis on Different GSA Methods Based on Monte Carlo Simulation

目的:探索差异表达基因集(DEGs)筛选的有效方法。方法:基于蒙特卡洛模拟比较Efron's GSA、SAFE、Globaltest、PCOT2等四种基因集方法在分析微阵列数据时的统计推断能力。结果:Globaltest和PCOT2两种基于模型构建的基因集方法在处理模拟微阵列数据时效能相当,Globaltest略优于PCOT2,而Efron's GSA、SAFE方法检验效能低下。结论:Globaltest是一种较有效的微阵列数据分析方法。

Objective：To investigate effective methods of identifying different expression gene set（DEGs）.Methods：The statistical inference capability of Efron＇s GSA,SAFE,Globaltest,PCOT2 methods,based on Monte Carlo simulation,were compared when used in microarray data analysis.Results：The efficiency of Globaltest and PCOT2 based on model building was almost the same in analyzing simulated microarry data,but Globaltest was slightly better than PCOT2.The detection rate of Efron＇s GSA and SAFE was lower.Conclusion：Globaltest is a more effective method to analysis microarray data.