Skip to main content


Figure 2 | Genome Biology

Figure 2

From: RELATCH: relative optimality in metabolic networks explains robust metabolic and regulatory responses to perturbations

Figure 2

Comparison of MFA estimated flux values and predicted flux values by different methods for four E. coli knockout mutants before and after adaptive evolution. (a) The MFA estimated (x-axis) and predicted (y-axis) flux values, sum of squared errors per flux (SSE) and Pearson's correlation coefficient (r) are shown for each unevolved mutant. Error bars indicate the confidence intervals from MFA. The RELATCH predictions were made using the tight parameter values. (b) SSE and r values for different approaches used to predict behaviors for evolved mutants, where the average MFA results (across two independently evolved strains) were used to calculate the SSE and r values. The RELATCH predictions were made using the relaxed parameter values. (c) Largest relative flux changes predicted by RELATCH in evolved mutant strains. The relative flux changes, determined as ratios of predicted mutant fluxes to wild-type fluxes, were rank-ordered to identify reactions with the largest changes, which were then compared to experimental data, including enzyme activity, gene expression [24], and genetic mutations [27]. AKG, 2-oxoglutarate; DHAP, dihydroxyacetone phosphate; G6P, glucose-6-phosphate; GLCEXT, extracellular glucose; ICIT, isocitrate; PEP, phosphoenolpyruvate; PYR, pyruvate; SUCC, succinate.

Back to article page