Fig. 6: Impact of multi-strain data integration on model accuracy.
From: Rational strain design with minimal phenotype perturbation
Anthranilate titers from a the 35 models calibrated on data from two strains – W3110 trpD9923 and W3110 trpD9923/pJLaroGfbr, and b the 13 models calibrated on all three strains, including W3110 trpD9923/pJLaroGfbtktA. Simulation results are denoted by solid lines (median) and shaded regions (interquartile range) for W3110 trpD9923 (orange), W3110 trpD9923/pJLaroGfbr (black), and W3110 trpD9923/pJLaroGfbrtktA (red). Results of the triplicate experiments are denoted by solid circles (mean) and error bars (standard deviation) for W3110 trpD9923 (circle), W3110 trpD9923/pJLaroGfbr (cross), and W3110 trpD9923/pJLaroGfbrtktA (triangle). Incorporating additional data improves accuracy in capturing the experimental behavior of all three strains. Source data are provided as a Source Data file.
