Associations between carotid artery intima-media thickness, traditional risk factors and proteins

Figures

• 

  Fig. 1. Workflow of data collection and analyses in this study. Datasets and variables used for analysis were detailed in this figure.

• 

  Fig. 2. Intima-media thickness (IMT) and CVD risk factors. Relationships between IMT of the carotid artery and measured levels (Obs) or genetically determined levels of the traditional CVD risk factors, including using polygenetic risk scores (PRS), and Mendelian randomization [inverse-variance weighted (IVW) and MR Egger (Egger)] techniques. The regression coefficient (beta) and 95%CI are shown.

• 

  Fig. 3. Intima-media thickness (IMT) and measured proteins. Relationships between measured protein levels and IMT. Only the 63 proteins with FDR < 0.05 in the age and sex-adjusted model and p < 0.05 in the multiple CVD risk factors-adjusted models are given in the figure. The estimates shown here are age, sex and fasting time-adjusted.

• 

  Fig. 4. Intima-media thickness (IMT) and proteins of interest by Mendelian randomization (MR). Relationships between the five proteins being significant in the MR analysis vs IMT and the corresponding results for these proteins in the observational analysis (Obs) adjusted for age and sex.

• 

  Fig. 5. Colocalization analysis of proteins of interest and intima-media thickness (IMT). The figure shows the posterior probability of five exclusive hypotheses to support whether proteins of interest and IMT share the same causal genetic variants in the defined genomic region.

• 

  Fig. 6. Genetic analyses of proteins of interest. A Expression analysis in three relevant arteries for genes of five proteins being significant in the Mendelian randomization (MR) analysis vs the intima-media thickness (IMT). B Heat-map of MR-Wald ratio estimate of selected proteins with traditional risk factors of IMT. Significant levels are expressed as ***, **, * or lack of * if nominal p-value is <0.001, <0.01, < 0.05 or > 0.05. A blank square indicates that no association was estimated because of lack of genetic instruments. C Mediation analysis of the effect of proteins on IMT via SBP using MR.