Identification and genetic validation of potential therapeutic targets for pulmonary hypertension through multi-omics causal inference

Figures

• 

  Figure 1.. Research workflow diagram. CD300LF = CD300 molecule like family member f, DAP = death-associated protein, deCODE = deCODE genetics, eQTL = expression quantitative trait loci, GWAS = genome-wide association study, MR = Mendelian randomization, PF4V1 = platelet factor 4 variant 1, PPH4 = posterior probability for hypothesis 4, pQTL = proteomic quantitative trait loci, SMR = summary-data-based Mendelian randomization.

• 

  Figure 2.. Screening of causal targets for PH through multilayered MR analysis. (A-C) Volcano plots showing causal effect estimates from (A) proteome-wide pQTL-MR, (B) eQTL-MR, and (C) SMR analyses. The x-axis represents the effect estimate on PH risk, and the y-axis indicates statistical significance (−log10 P-value). (D) Summary of consistent causal effects (odds ratio, 95% CI, and P-value) for the 6 final MR-ICTs across all 3 MR methods. CI = confidential interval, eQTL = expression quantitative trait loci, ICT = information and communication technology, MR = Mendelian randomization, OR = odds ratio, PH = pulmonary hypertension, SMR = summary-data-based Mendelian randomization.

• 

  Figure 3.. Bayesian colocalization analysis of pulmonary hypertension risk loci. LocusZoom plots showing colocalization analysis at the (A) LYZ, (B) PF4V1, (C) GREM2, and (D) NID1 loci. The y-axis represents the association P-value with PH (−log10 P). The alignment of top signals for both the molecular trait and PH at each locus supports a shared causal variant. SNP = single nucleotide polymorphism.

• 

  Figure 4.. Functional enrichment analysis of MR-ICTs. (A) Dot plot of GO enrichment analysis. The y-axis represents GO terms, and the x-axis represents gene ratio. The color of the dots indicates the significance of the adjusted P-value (redder indicates higher significance), and the size of the dots represents the number of genes enriched in that term. (B) Gene-concept network plot of GO enrichment analysis. Yellow nodes represent significantly enriched GO terms, while other colored nodes represent MR-ICTs genes. Edges indicate the affiliation between genes and corresponding GO terms. (C) Dot plot of KEGG pathway enrichment analysis. The meanings of parameters (color, size, x-axis) are consistent with (A). (D) Network plot of KEGG enrichment analysis. Yellow nodes represent significantly enriched KEGG pathways, while other colored nodes represent MR-ICTs genes.

• 

  Figure 5.. Cell type-specific expression profiles of MR-ICTs in human peripheral blood mononuclear cells. (A) t-SNE (t-distributed stochastic neighbor embedding) dimensionality reduction visualization plot generated from single-cell RNA-sequencing data. (B-G) Feature plots illustrating the expression levels of the 6 MR-ICTs encoding genes: (B) LYZ, (C) DAP, (D) PF4V1, (E) CD300LF, (F) GREM2, and (G) NID1, across the t-SNE cell map. Color gradients represent normalized gene expression levels, ranging from gray (low expression) to red (high expression). (H-M) Violin plots quantifying and displaying the expression distribution of these 6 genes (H-M) Within each annotated cell subpopulation.

• 

  Figure 6.. PheWAS study of MR-ICTs to assess potential pleiotropy. Manhattan plots illustrating the gene-level association results of MR-ICTs with various phenotypes in the UK Biobank. For each gene: LYZ (A, B), DAP (C, D), PF4V1 (E, F), CD300LF (G, H), GREM2 (I, J), and NID1 (K, L), plots (A, C, E, G, I, K) show associations with binary traits, and plots (B, D, F, H, J, L) show associations with continuous traits. In the plots, the y-axis represents the statistical significance of association (−log10 [P-value]), and the x-axis represents clinical phenotypes grouped by category. The horizontal black dashed line indicates the genome-wide significance threshold (P = 5 × 10−8).