双样本双向孟德尔随机化法分析人免疫细胞与增生性瘢痕之间的因果关系  

Analysis of the causal relationship between human immune cells and hypertrophic scar using two-sample bidirectional Mendelian randomization method

作　　者：吴虹林 陈咏菲 李舒婷[2] 杨浩 李晓卉[1] 唐冰[1] 朱家源[1] 胡志成[1] Wu Honglin;Chen Yongfei;Li Shuting;Yang Hao;Li Xiaohui;Tang Bing;Zhu Jiayuan;Hu Zhicheng(Department of Burn and Wound Repair Surgery,the First Affiliated Hospital of Sun Yat-sen University,Guangzhou 510080,China;Department of Plastic Surgery,the First Affiliated Hospital of Sun Yat-sen University,Guangzhou 510080,China)

机构地区：[1]中山大学附属第一医院烧伤与创面修复科,广州510080 [2]中山大学附属第一医院整形外科,广州1510080

出　　处：《中华烧伤与创面修复杂志》2024年第6期572-578,共7页Chinese Journal of Burns And Wounds

基　　金：国家自然科学基金面上项目(82172213)。

摘　　要：目的运用双样本双向孟德尔随机化法,探讨人免疫细胞与增生性瘢痕(HS)之间的因果关系。方法该研究基于双样本双向孟德尔随机化(MR)法,分别从全基因组关联分析catalog数据库和FinnGen数据库获取731种人免疫细胞与HS的数据集。设置显著阈值,筛选与免疫细胞及HS显著相关的单核苷酸多态性(SNP)并排除弱工具变量偏倚的影响。采用逆方差加权(IVW)法[同时采用错误发现率(FDR)中的Benjamini-Hochberg法校正其P值]初步检测免疫细胞与HS的因果关系并筛选与HS有显著因果关系的免疫细胞。进一步地,采用双样本MR分析的5种方法:IVW法、加权中位数法、简单模式法、加权模式法和MR-Egger法绘制散布图检测筛选出的免疫细胞与HS的因果关系。针对符合假设的免疫细胞SNP,进行Cochran Q检验评估异质性,进行MR-Egger回归和MR-PRESSO法排除水平多效性,进行留一法分析来评估显著结果是否由单个SNP决定。采用双样本MR分析中IVW法反向检测HS与免疫细胞之间的因果关系。结果731种免疫细胞达到显著阈值的SNP数量从7个到1786个不等,HS中达到显著阈值的SNP数量有119个,这些SNP的F值均>10,提示具有弱工具变量偏倚的可能性较小。IVW法分析显示,60种免疫细胞与HS具有潜在因果关系(P值均<0.05),经Benjamini-Hochberg法校正后,仅CD45RA和CD39双阳性调节性T细胞(Treg)与HS具有潜在因果关系(P FDR<0.05)。IVW法(比值比为1.16,95%置信区间为1.08~1.24,P<0.05、P FDR<0.05)、加权中位数法(比值比为1.16,95%置信区间为1.05~1.28,P<0.05)、加权模式法(比值比为1.14,95%置信区间为1.02~1.27,P<0.05)和MR-Egger法(比值比为1.18,95%置信区间为1.07~1.30,P<0.05)的散布图均显示,CD45RA和CD39双阳性Treg的14个SNP与HS患病风险存在因果关系;仅简单模式法的散布图显示,CD45RA和CD39双阳性Treg的14个SNP与HS患病风险的因果关系不明显(P>0.05)。Cochran Q检验显示,CD45RA和CD39�ObjectiveTo explore the causal relationship between human immune cells and hypertrophic scar(HS)using two-sample bidirectional Mendelian randomization(MR)method.MethodsThis study was based on two-sample MR method,and the datasets of 731 immune cells and HS were obtained from the genome-wide association study(GWAS)catalog database and Finngen database,respectively.A significance threshold was established to discern single nucleotide polymorphism(SNP)significantly correlated with immune cells or HS,thereby eliminating the impact of weak instrumental variable bias.The inverse variance weighted(IVW)method(meanwhile,the Benjamini-Hochberg(BH)procedure of false discovery rate(FDR)to adjust P values)was used for preliminary detection of the causal relationship between immune cells and HS and screen the immune cells that had a significant causal relationship with HS.Further,the causal relationship between the selected immune cells and HS was detected through five two-sample MR methods:IVW method,weighted median method,simple mode method,weighted mode method,and MR-Egger method,and the scatter plot was drawn.SNPs conformed to the hypothesis were subjected to Cochran Q test for heterogeneity assessment,MR-Egger regression coupled with MR-PRESSO to eliminate horizontal pleiotropic effects,and a leave-one-out analysis was also conducted to determine if significant results were driven by individual SNP.Finally,the IVW method contained in the two-sample MR analysis was utilized to inversely examine the causal relationship between HS and immune cells.ResultsThe number of SNPs in 731 immune cells reaching the significance threshold varied from 7 to 1786,while in HS,119 SNPs met the significance threshold,with the F values of all SNPs being greater than 10,suggesting a low likelihood of bias from weak instrumental variables.The IVW method revealed that 60 types of immune cells potentially had a causal relationship with HS(with all P values<0.05),and after adjustment using the BH method,only CD45RA and CD39 positive regulatory T c

关 键 词：免疫 细胞 孟德尔随机化分析 因果律 增生性瘢痕 

分 类 号：R622[医药卫生—整形外科]