孟德尔随机化研究及其在中医药领域的应用展望

Mendelian Randomization and Its Application in Traditional Chinese Medicine

孟德尔随机化研究以遗传变异作为工具变量建立模型,用于测试风险因素与各种表型间的因果关系,目前已被广泛用于流行病学中疾病复杂病因的探索。孟德尔随机化研究能有效克服混杂和反向因果导致的偏倚,且易于实施,使其在中医药领域进行因果关系的推断变得可行。通过对孟德尔随机化研究的概念、核心假设、常见类型、在医学领域的研究现状及当前面临的弱工具变量、多效性、连锁不平衡等问题的阐述和分析认为,(1)孟德尔随机化研究在中医药领域具有较大的应用发展空间,其有助于中医证型的简化,理清不同病证间的关系,实现“异病同治”的目的;(2)该方法可有效补充观察性研究和干预性研究的不足,为中医药研究提供新方法;(3)同时存在可用数据相对局限的问题。

Mendelian randomization uses genetic variants as instrumental variables to establish the testing model for the causal relationship between risk factors and various phenotypes, and has been widely used in exploring the complex etiology of diseases in epidemiology. This method can effectively overcome the bias caused by confounding and reverse causation, and is easy to implement, making it feasible to infer causality in the field of traditional Chinese medicine(TCM). Through analysis on the concept, core assumptions, common types, the medical research status of Mendelian randomization, as well as the current challenges such as weak instrumental variables, pleiotropy, and linkage disequilibrium, it is believed that there is spacious room for Mendelian randomization in TCM researches. With it, the TCM syndrome types could be simplified, and the relationship between different diseases and syndromes may be clarified, thereby making “many different diseases can be treated in the same way” a reality. Meanwhile, Mendelian randomization can overcome certain limitations of the observational and interventional studies, and may provide new method for the exploration of TCM. However, there is a problem of relatively limited available data.