||Observational studies are useful to identify associations between risk factors and diseases, but often show inconsistent results due to heterogeneity among studies such as metrics, race, and cohort characteristics. In addition, many associated factors have a bidirectional interaction with the disease. For example, although the risk of disease is increased by an associated factor, it can be also increased due to the occurrence of the disease. In the case of such a complex epidemiologic relationship, it is difficult to confirm the exact causality or direction of interaction only by observational studies. The concept of causality is important in medicine because it enables proper and effective treatments. Randomized controlled trials (RCTs) is considered a gold standard to confirm a causal relationship between risk factors and disease. For example, in interventional studies such as drug studies, causality can be confirmed through RCT. In many cases such as smoking and lung cancer, however, RCTs are not feasible because it is unethical to induce risk factors through intervention. The Mendelian randomized study (MR) is based on a fact that genetic variants which increase or decrease the risk of a factor are randomly assigned during meiosis. Comparing the difference in the risk of disease for two groups that have high or low genetic risk for the factor can evaluate whether the factor induces the disease without intervention. In this presentation, I will talk about genome-wide associations studies that identify genetic variants associated with traits such as risk factors and diseases. Then, I will explain MR analyses, assumptions, interpretation as well as several examples from literature.