摘要
目的 在探讨基因与疾病关系的研究中比较不同设计所需的样本量。方法 依据logistic回归估计相对危险性值的原理 ,以模拟参数值回代计算样本量 ,计算过程以QUANTO软件完成。结果 (1)在基因间存在交互作用时 ,人群中基因易感性频率 (Pr)大于 2 0 % ,四种设计方法因样本量低于 10 0 0 ,可行性较大 ,以Pr为 3 5 %时样本量最小 :病例—父母设计为首选 ,单纯病例设计仅可用于基因间交互作用的初筛 ,但不能分析基因的主作用。 (2 )基因间不存在交互作用时 ,首选病例—对照设计。结论 两基因交互作用研究所需样本量的大小取决于不同的研究设计。
Objective To investigate sample size requirements for studies of gene-gene(G×G) interaction focusing on four study designs: the matched-case-control design,the case-sibling design,the case-parent design,and the case-only design.Methods Selecting different parameters (such as inheritance mode,susceptibility frequency,frequency of allele for disease,OR of gene main effect) to calculate sample sizes on QUANTO software based on logistic regression model.Results (1)The case-parent design was the best method to choose when there was interaction between genes. (2)The matched-case-control design was the best mean for studying the main effect of genes when there were no interaction .(3)It is difficult to carry out a research when susceptibility frequency of the genes were less than 20% because the size of sample were bigger than 1000. Conclusion The size of sample for research the interaction effect of two genes dependences on types of design.
出处
《中国卫生统计》
CSCD
北大核心
2003年第6期341-343,共3页
Chinese Journal of Health Statistics
