Genetic correlation structure of established endocrine proteins.

A, All genes across the 18 metabolic tissues in 310 individuals were correlated with expression of ADIPOQ in subcutaneous adipose tissue, where a qvalue cutoff of q<0.1 showed the strongest enrichments with local subcutaneous and muscle gene expression (pie chart, left). The top 500 genes which correlated with subcutaneous ADIPOQ were used for an overrepresentation test across Gene Ontology Biological process annotations, where pathways related to fatty acid oxidation were observed in adipose (middle) and ribosome/metabolic processes in skeletal muscle (right). B-D, The same qvalue binning, local and top peripheral enrichments were applied to intestinal GCG (B), liver FGF21 (C) and muscle IL6 (D). For these analyses all 310 individuals (across both sexes) were used and qvalue adjustments calculated using a Benjamini-Hochberg FDR adjustment.

Genetic correlation structure and network architecture of liver PCSK9 and adipose PNPLA2.

A, distribution of pan-tissue genes correlated with liver PCSK9 expression (q<0.1), where 93% of genes were local in liver (purple). B, Gene ontology (BP) overrepresentation test for the top 500 hepatic genes correlated with PCSK9 expression in liver. C, Undirected network constructed from liver genes (aqua) correlated with PCSK9, where those annotated for “cholesterol biosynthetic process” are colored in red. D-E, over-representation tests corresponding to the top-correlated genes with adipose (subcutaneous) PNPLA2 expression residing locally (D) or peripherally in skeletal muscle (E). F, Undirected network constructed from from the strongest correlated subcutaneous adipose tissue (light aqua) and muscle genes (dark blue) with PNPLA2 (black), where genes corresponding to GO terms annotated as “fatty acid beta oxidation” or “Muscle contraction” are colored purple or red, respectively. For these analyses all 310 individuals (across both sexes) were used and qvalue adjustments calculated using a Benjamini-Hochberg FDR adjustment. Network graphs generated based in Biweight midcorrelation coefficients, where edges are colored blue for positive correlations or red for negative correlations.

Pan-tissue circadian clock concordances highlights centrality of colon ARNTL.

A Heatmap showing Biweight midcorrelation coefficients between key circadian rhythm genes across metabolic tissues. Positive correlations are shown in purple negative correlations in green, where a * indicates significance (regression Pvalue<0.01). B, Frequency (y-axis) of circadian rhythm genes per tissue (x-axis), where regression P value was less than 0.001 across all other rhythm genes in other tissues. C, The top-ranked rhythm genes (x-axis) based on biweight midcorrelation (y-axis) with expression of ARNTL in colon. D-E, The top-ranked cell types (x-axis) correlated with colon ARNTL expression based on Biweight midcorrelation coefficient (y-axis), either locally within colon (D), or across peripheral tissue deconvoluted cell abundances (E). For these analyses all 310 individuals (across both sexes) were used and regression pvalues calculated from the bicor coefficient using WGCNA41.

Putative muscle-hippocampal brain signals and web tool overview.

A, Genes encoding secreted proteins (x-axis) plotted against the average -log(regression-pvalue) with all transcripts in hippocampus (y-axis), termed Ssec score5. B-G, For select genes shown in A, the pathway enrichments from overrepresentation tests based on the top 500 genes within skeletal muscle (B, D, F) or Hippocampus (C, E, G) are shown for ADAMTS17 (B-C), FNDC5 (D-E) or ERFE (F-G). H, Schematic for design of web portal to query all gene-gene and gene-cell genetic correlations and run targeted analyses.