Kou Ni, Zhou Wenyang, He Yuzhu, Ying Xiaoxia, Chai Songling, Fei Tao, Fu Wenqi, Huang Jiaqian, Liu Huiying
Department of Oral Prosthodontics, School of Stomatology, Dalian Medical University, Dalian, China.
School of Life Science and Technology, Harbin Institute of Technology, Harbin, China.
Front Bioeng Biotechnol. 2020 Mar 20;8:201. doi: 10.3389/fbioe.2020.00201. eCollection 2020.
Accumulating evidence showed that Interleukin (IL) level is associated with Osteoporosis. Whereas, most of these associations are based on observational studies. Thus, their causality was still unclear. Mendelian randomization (MR) is a widely used statistical framework that uses genetic instrumental variables (IVs) to explore the causality of intermediate phenotype with disease. To classify their causality, we conducted a MR analysis to investigate the effect of IL-18 level on the risk of Osteoporosis. First, based on summarized genome-wide association study (GWAS) data, 8 independent IL-18 SNPs reaching genome-wide significance were deemed as IVs. Next, Simple median method was used to calculate the pooled odds ratio (OR) of these 8 SNPs for the assessment of IL-8 on the risk of Osteoporosis. Then, MR-Egger regression was utilized to detect potential bias due to the horizontal pleiotropy of these IVs. As a result of simple median method, we get the SE (-0.001; 95% CI-0.002 to 0; = 0.042), which means low IL-18 level could increases the risk of the development of Osteoporosis. The low intercept (0; 95% CI -0.001 to 0; = 0.59) shows there is no bias due to the horizontal pleiotropy of the IVs.
越来越多的证据表明，白细胞介素（IL）水平与骨质疏松症有关。然而，这些关联大多基于观察性研究。因此，它们之间的因果关系仍不明确。孟德尔随机化（MR）是一种广泛使用的统计框架，它使用基因工具变量（IVs）来探索中间表型与疾病之间的因果关系。为了明确它们之间的因果关系，我们进行了一项MR分析，以研究IL-18水平对骨质疏松症风险的影响。首先，基于汇总的全基因组关联研究（GWAS）数据，将8个达到全基因组显著性的独立IL-18单核苷酸多态性（SNPs）视为IVs。接下来，使用简单中位数法计算这8个SNPs的合并比值比（OR），以评估IL-8对骨质疏松症风险的影响。然后，利用MR-Egger回归检测由于这些IVs的水平多效性导致的潜在偏差。通过简单中位数法，我们得到的SE为（-0.001；95%CI为-0.002至0；P = 0.042），这意味着低IL-18水平可能会增加患骨质疏松症的风险。低截距（0；95%CI为-0.001至0；P = 0.59）表明不存在由于IVs的水平多效性导致的偏差。
