Extended Data Fig. 3. Differential Liver Gene Expression, Pathway Enrichment, and Plasma Metabolomics Across Cardiometabolic and Liver-Specific Clusters.
Upper volcano plots illustrate the comparative liver gene expression in (a) the cardiometabolic cluster (CM; n = 97) vs control cluster (CTRL; n = 671), (b) the liver-specific cluster (LS; n = 63) vs control cluster, and (c) the cardiometabolic cluster vs liver-specific cluster. Horizontal and vertical lines represent adjusted p-value with Benjamin Hochberg correction (0.05) and log2FoldChange (0.26) significance thresholds respectively. Bar plots representing Gene Ontology Biological Processes (GO-BP) enrichment analysis for differentially expressed genes in the liver between (d) the cardiometabolic cluster (CM; n = 97) vs the control cluster (CTRL; n = 671), (e) the liver-specific cluster (LS; n = 63) vs the control cluster, and (f) the cardiometabolic cluster vs the liver-specific cluster. Graphs were generated using the GSEA enrichment method from the R package clusterProfiler. Lower volcano plots illustrate the comparative metabolomics in (g) cardiometabolic cluster (CM; n = 151) vs control cluster (CTRL; n = 1076), (h) liver-specific cluster (LS; n = 95) vs control cluster, and (i) cardiometabolic cluster vs liver-specific cluster. Horizontal and vertical lines represent adjusted p-value of 0.05 with Benjamini-Hochberg correction and log2FoldChange (0.26) significance thresholds respectively. The statistical tests are based on Limma’s linear model, performing a two-sided moderated t-test for each variable. P-values are adjusted for multiple comparisons using the Benjamini-Hochberg method to control the False Discovery Rate (FDR).