Table 1:

Methods to address pleiotropy in MR studies

Method	Description	Advantages	Disadvantages	Reference
MR Egger	Provides unbiased estimates of the true effect even in the presence of pleiotropy.	Does not require the instrumental variables to be valid.	Relies on the InSIDE (INstrument Strength Independent of the Direct Effect) assumption.	[25]
Multivariable MR analysis	Allows for simultaneous assessment of multiple correlated risk factors with a shared genetic basis, such as lipids.	Allows for inclusion of pleiotropic variants when the variants are associated with the risk factors under study.	Subject to the same assumptions as traditional MR analysis, including no pleiotropic associations with any factors that are not considered in the primary analysis.	[26]
Weighted median estimator	Calculates a weighted median of the individual SNP-specific estimates.	Does not rely on the InSIDE assumption.	Requires that at least 50% of the instrument variables are valid.	[27]
MR-PRESSO	Corrects for horizontal pleiotropic effects by removing outliers.	MR-PRESSO can be used in the context of other MR approaches such as MR Egger	Sensitive to the InSIDE assumption. Requires that at least 50% of the instrument variables are valid. Requires simulations, and thus is not as computationally fast as other methods.	[28]