Fig. 6: Interrogating controller performance under complex perturbations.

The first column presents the scaled disturbance profile, which is the dynamical deviation of the test group from the intact group, scaled by each construct's pre-disturbance output value. Data in this column are presented as mean value +/− standard deviation of n samples, n = number of biological replicates. The second and third columns present the robustness and the speed of the four network designs. Data from these two columns are presented with standard box plots overlaying all individual data points. Box plots show centre line as median, box limits as upper and lower quartiles, whiskers as minimum to maximum values. The asterisks denote statistical significance between the two groups, determined by paired one-tail Student's t-tests, additional statistical information is listed in Supplementary Table 8. If a trajectory does not recover back to the error band at the end of the measurement, a high-order polynomial function was then used to fit the data and predict its settling time. The asterisks denote statistical significance between the two groups. a System dynamics under chemical perturbation (spike), where the inducers AHL-Cin increased from 3 μM to 10 μM, and the AHL-Rhl increased from 20 μM to 100 μM at 310 min. Subsequently, the cultures were restored to the pre-disturbance media after 2.5 h. b System dynamics under temperature perturbation (dip). The temperature drops from 37 °C to 30 °C at 380 min and restores to 37 °C at 510 min. c System dynamics under temperature perturbation (spike). The temperature increases from 37 °C to 42 °C at 380 min and restores to 37 °C at 510 min. d System dynamics under nutrient perturbation (spike). The glucose concentration spikes from 0.1% to 1% at 370 min and restores to 0.1% at 500 min. e System dynamics under nutrient perturbation (dip). The glucose concentration drops from 1% to 0% at 370 min and restores to 1% at 500 min.