Fst统计法评估KIF1B基因rs17401966多态性与肝细胞癌风险Meta分析的异质性来源

目的采用Fst统计法评估二分类变量Meta分析的异质性来源。方法检索Google Scholar、EMBASE、PubMed、ISI Web of Knowledge和CNKI数据库,搜集有关KIF1B基因rs17401966多态性与肝细胞癌风险关系的病例-对照研究。应用Stata 12.0软件进行Meta分析,使用Arlequin 3.5软件分析研究人群间的遗传分化程度。结果最终纳入5个病例-对照研究,合计12个研究人群。对12个人群的Meta分析结果显示KIF1B基因rs17401966G等位基因与HCC风险负相关(OR=0.77,95%CI:0.65~0.93;P=0.005),然而合并结果存在很强的异质性。通过对12个纳入人群进行遗传分化检验,发现Zhang等的5个研究人群与其他研究人群存在不同程度的遗传分化,进而根据Fst值进行适当的亚组分析,把组8和组9异质性检验的I2值降到了25%以下,然而组8和组9的Meta分析结果却不一致。结论研究显示在对KIF1B基因rs17401966多态性与肝细胞癌风险进行Meta分析时,通过对纳入人群的遗传分化检验,可以发现Meta分析的异质性来源。

Fst statistical method for evaluating KIF1B gene rs17401966 polymorphism and heterogeneity source of hepatocellular carcinoma risk meta-analysis

Objective To adopt Fst statistical method to assess the heterogeneity sources of meta‐analysis by dichotomous varia‐ble .Methods The case‐control studies on the relationship between KIF1B gene rs17401966 polymorphism and hepatocellular carci‐noma(HCC) risk were collected by retrieving the databases including Google Scholar ,EMBASE ,PubMed ,ISI Web of Knowledge and CNKI .The meta analysis was performed by using the Stata12 .0 software .The genetic differentiation degree among populations was analyzed and researched by using the Arlequin 3 .5 software .Results A total of 5 case‐control studies were finally included ,in‐volving 12 research populations .The meta analysis on 12 populations showed that KIF1B gene rs17401966G allele was negatively correlated with HCC risk (OR=0 .77 ,95% CI:0 .65-0 .93 ;P=0 .005) .However ,the strong heterogeneity existed in this pooled re‐sults .The genetic differentiation test in the included 12 populations found that Zhang′s five research populations had varying de‐grees of genetic differentiation compared to other populations .Then the proper subgroup analysis was further conducted based on Fst value ,and then the I2 value of the heterogeneity test in the group 8 and 9 was descended to less than 25% .However ,the meta analysis results of the group 8 and 9 were inconsistent .Conclusion This study showed that conducting the meta‐analysis of KIF1B gene rs17401966 polymorphism and the HCC risk can find the heterogeneity sources of meta‐analysis by conducting the genetic dif‐ferentiation test in the included population .