Fig. 4: Clustering analysis of stable state solutions.

A The primary heatmap displays four clusters of data derived from different Myc values (0, 300 nM, 1200 nM, 3000 nM). Each row in this heatmap represents a stable-state solution, and each column represents one variable in the model. The proportions of each Myc group are shown in the first one-column heatmap, with a summary annotation at the top presenting the proportion of each Myc level within each cluster in a boxplot format (from left to right: “Q”, “W”, “O”, and “W/O”). The second one-column heatmap, labeled ‘Metabolic State’, corresponds to rows indicating whether steady states are active or inactive based on their glucose oxidation, glycolysis, and fatty acid oxidation values having Z-scores greater or less than 0, respectively. The third one-column heatmap shows the glucose uptake rates and the average for each cluster on the top, units expressed in µM/s. The subsequent one-column heatmaps, labeled ‘ATP Prod’, ‘ATP Cons’, and ‘ATP Overall’, illustrate the total ATP production, consumption, and net ATP for each row, taking into account both anabolic and catabolic processes. B A plot of pAMPK vs HIF-1 illustrates the relationship between pAMPK and HIF-1 levels across all generated steady states. C A scatter plot showing the results of a principal component analysis (PCA). Each point in the plot represents a steady state, plotted according to its scores on the first two principal components (PC1 and PC2) of the output element values in the minimal regulatory network. Different clusters are indicated by different colors, providing a visual representation of the grouping of steady states in the reduced-dimensional space of the PCA. Both plots (B, C) show the state annotations for the ‘L/L’ cluster, indicating the position of the Low Metabolic State. These are considered low due to their glucose oxidation, glycolysis, and Fatty acid Oxidation values having Z-scores less than 0, respectively. Bar plots showcasing the loading variables PC1(D) and PC2 (E) derived from the PCA. The variables are ranked based on their loading values, signifying their contribution to each principal component. Variable abbreviations: glutathione synthase (GSH), HIF-1 (h), glycolysis (G₂), noxROS (R_nox), glucose uptake (G₀), reductive fatty acids (F_re), reductive glucose (G_re), glutamine uptake (Q₀), fatty acids uptake (F₀), pAMPK (A), rate of acetyl-CoA for mitochondrial respiration (C₀), FAO (F), reductive glutamine (Q_re), glucose oxidation (G₁), mitochondrial ROS (R_mt), glutamine oxidation (Q₁).