Fig. 3: Tumour versus normal changes in metabolite and transcript abundance are predominantly asynchronous.
From: A multimodal atlas of tumour metabolism reveals the architecture of gene–metabolite covariation
a, Heat map of metabolite and transcript DA scores across datasets and pathways, capturing the tendency for metabolites and genes to accumulate or deplete in tumours relative to normal tissues. The size of the dots indicates the number of molecules measured in that pathway, while the colour represents the DA score. b, Spearman correlation coefficients of the metabolite and transcript DF scores in KEGG pathways. Red dots indicate nominal significance (P value < 0.05). A minority of pathways showed significant association between metabolomic and transcriptomic disruption across CAMP datasets. P values were estimated from Spearman’s rank correlation test and were corrected for multiple testing using the Benjamini–Hochberg method. c, Example of Spearman correlation calculation: Distribution of the metabolite (x axis) and transcript (y axis) DF scores across datasets for the citrate cycle (tricarboxylic acid (TCA) cycle) pathway. P values were estimated from Spearman’s rank correlation test.
