Integrative Cross-platform Analysis of Kinase Inhibitor Effects on Statin-relevant Cardioprotective Pathways in Human Cardiomyocytes

Figures

• 

  Figure 1. Study design and analytical workflow. Six statin-relevant cardioprotective gene sets were analyzed independently in two transcriptomic datasets and by cell-line-adjusted proteomics. Integration was pathway-specific; the invalid unequal-universe per-KI cross-omics mean was not retained; common-five estimates are descriptive sensitivity analyses. KI, Kinase inhibitor; NES, normalized enrichment score; GSEA, gene set enrichment analysis.

• 

  Figure 2. Discovery GSEA of kinase inhibitor effects on statin cardioprotective pathways. Heatmap of NES from fgsea (20) for 23 kinase inhibitors across 6 statin pathways in GSE146096. Red, positive NES; blue, negative NES; asterisks denote global FDR. *FDR<0.05, **FDR<0.01, and ***FDR<0.001. GSEA, Gene set enrichment analysis; NES, normalized enrichment score; FDR, false discovery rate.

• 

  Figure 3. Signed pathway NES across discovery and validation transcriptomic datasets. (A) GSE146096. (B) GSE217421. Red denotes positive NES and blue denotes negative NES. Colors indicate enrichment direction within each kinase inhibitor dataset and are not interpreted as statin alignment or opposition. NES, Normalized enrichment score. *FDR<0.05, **FDR<0.01, ***FDR<0.001.

• 

  Figure 4. Pathway-specific cross-dataset transcriptomic concordance. Each panel compares KI NES between GSE146096 and GSE217421. Nominal p-values and Benjamini-Hochberg q-values across six pathways are shown. Anti-fibrotic concordance was ρ=0.495, p=0.016, q=0.098. KI, Kinase inhibitor; NES, normalized enrichment score.

• 

  Figure 5. Corrected pathway-specific transcriptomic-proteomic integration. (A) Cell-line-adjusted protein pathway scores for 21 KIs; NF-κB is unavailable because only 2/20 proteins were detected. (B) Anti-fibrotic concordance (ρ=0.644, p=0.0016, q=0.0081). (C) Cell-adjusted correlations and Benjamini-Hochberg q values for the five assessable pathways. The invalid unequal-universe per-KI mean is not shown; common-five per-KI estimates are reported only as a descriptive sensitivity analysis in Supplementary Table XIX. KI, Kinase inhibitor; NES, normalized enrichment score.

• 

  Figure 6. Exploratory kinase-target analyses. (A) Binary kinase-binding heatmap for 20 selected key targets (Kd<1 μM). (B) Total bound kinase-target count versus mean pathway NES (ρ=−0.378, p=0.076). (C) Exploratory MAPKAPK2 association with anti-fibrotic NES (discovery p=0.032; validation p=0.039); no kinase-pathway test survived FDR correction.