EWAS meta-analysis of frailty in two European twin cohorts

EWAS summary statistics for the paper "Epigenome-wide analysis of frailty: results from two European twin cohorts" by Mak et al. (2024).<b>Method:</b><b> </b>We first performed four cross-sectional EWASs of the frailty index (FI) in SATSA 450K, SATSA EPIC, LSADT 1997, and LSADT 2007 samples using generalized estimating equations where DNA methylation levels of CpG sites (i.e., normalized <i>β</i> values, representing proportion of methylation) were used as the dependent variables and FI as the independent variable. The models were adjusted for age, sex, smoking, and body mass index (BMI). Both twins from each pair were included in the models, and we accounted for twin relatedness using cluster-robust standard errors. The models in LSADT were additionally adjusted for batch and estimated cellular compositions (CD4 cells, CD8 cells, natural killer cells, monocytes, granulocytes, and B cells). We considered the FI both as a continuous variable (per 0.1 increase) and a 3-group categorical variable (frail vs. non-frail and prefrail vs. non-frail). Afterwards, a fixed-effect inverse variance weighted meta-analysis was performed using the metafor R package. The meta-analysis included 368,249 CpGs that were available in the 450K samples (i.e., SATSA 450K, LSADT 1997, and LSADT 2007). Potential heterogeneity was assessed using the <i>I</i>² value. CpGs associated with either the continuous or categorical FI at a false discovery rate (FDR) &lt;.05 were considered as statistically significant.<b>List of EWAS datasets:</b>"<i>EWAS_FI_cont_summary_stat.txt</i>": EWAS summary statistics of the continuous FI"<i>EWAS_frail-vs-nonfrail_summary_stat.txt</i>": EWAS summary statistics of the categorical FI (frail vs. non-frail)"<i>EWAS_prefrail-vs-nonfrail_summary_stat.txt</i>": EWAS summary statistics of the categorical FI (prefrail vs. non-frail)<br>

本数据集为Mak等人2024年发表的论文《衰弱的全表观基因组关联分析：两项欧洲双生子队列研究结果》（Epigenome-wide analysis of frailty: results from two European twin cohorts）的全表观基因组关联研究（Epigenome-Wide Association Study, EWAS）汇总统计数据。**研究方法**：本研究首先在SATSA 450K、SATSA EPIC、LSADT 1997及LSADT 2007队列样本中，针对衰弱指数（Frailty Index, FI）开展4项横断面全表观基因组关联研究，采用广义估计方程（Generalized Estimating Equations, GEE）进行建模：以CpG位点的DNA甲基化水平（即标准化β值，代表甲基化比例）作为因变量，衰弱指数作为自变量。所有模型均校正了年龄、性别、吸烟状态及体质量指数（Body Mass Index, BMI）。本研究纳入每对双生子的两名个体参与模型拟合，并采用聚类稳健标准误控制双生子间的亲缘关联混杂。LSADT队列的分析模型额外校正了批次效应及推定的细胞组成（CD4细胞、CD8细胞、自然杀伤细胞、单核细胞、粒细胞及B细胞）。本研究针对衰弱指数，分别以连续变量形式（每0.1单位增量）及3分类变量形式（衰弱组vs. 非衰弱组、衰弱前期组vs. 非衰弱组）进行分析。随后，采用metafor R包开展固定效应逆方差加权荟萃分析。本次荟萃分析纳入了450K队列样本（即SATSA 450K、LSADT 1997及LSADT 2007）中覆盖的368,249个CpG位点。采用I²值评估潜在的异质性。以错误发现率（False Discovery Rate, FDR）<0.05作为统计学显著性阈值，将与连续型或分类型衰弱指数存在关联的CpG位点判定为具有统计学显著性。**EWAS数据集列表**：<i>EWAS_FI_cont_summary_stat.txt</i>：连续型衰弱指数的EWAS汇总统计数据；<i>EWAS_frail-vs-nonfrail_summary_stat.txt</i>：分类型衰弱指数（衰弱组vs. 非衰弱组）的EWAS汇总统计数据；<i>EWAS_prefrail-vs-nonfrail_summary_stat.txt</i>：分类型衰弱指数（衰弱前期组vs. 非衰弱组）的EWAS汇总统计数据。