Using Mendelian randomization analysis to determine the causal connection between unpleasant emotions and coronary atherosclerosis

Figures

• 

  Figure 1. Design framework and assumptions of the Mendelian randomization study. Independence assumption, the genetic variants chosen as instrumental variables(IVs) should not be linked to any confounding factors; correlation assumption, IVs should be significantly associated with the exposure; exclusivity assumption, IVs should only impact the risk of an outcome via the exposure's risk factor, and not through alternate routes. SNPs, single nucleotide polymorphisms.

• 

  Figure 2. Forest plot of three Mendelian random estimators of the effect of unpleasant emotions on coronary atherosclerosis. IVW, indicates inverse variance weighted; MR, Mendelian randomization; OR, odds ratio.

• 

  Figure 3. Scatterplot of Mendelian randomization estimates of genetic risk of unpleasant emotions on coronary atherosclerosis. Each point represents the SNP effects on unpleasant emotions and coronary atherosclerosis. The line at each point reflects the 95% confidence interval, the horizontal coordinate is the effect of SNP on exposure (unpleasant emotions) and the vertical coordinate is the effect of SNP on the outcome (coronary atherosclerosis). SNPs, single nucleotide polymorphisms.

• 

  Figure 4. Leave-one-out analysis of unpleasant emotions on coronary atherosclerosis. Circles indicate MR estimates for unpleasant emotions and coronary atherosclerosis using the inverse-variance weighted fixed-effect method if the SNP was omitted. The bars indicate the CI of MR estimates. CI, confidence interval; MR, Mendelian randomization.

• 

  Figure 5. Funnel plot of the Mendelian randomization analysis for coronary atherosclerosis. The funnel plot assessed the existence of probable heterogeneity among the estimations, which suggests the possibility of pleiotropic effects. The graph depicts the observed causal impact of each of the 40 instrumental variables (IVs) as dots, and the average causal effect of all IVs combined (bIV) using the inverse variance weighted and MR-Egger approach on the x-axis. The Y axis indicates the inverse standard error of the predicted causal impact for each single nucleotide polymorphism (SNP).