Fig. 4: Computation of controllability based on the entropies for the different cell systems and cultivation devices used.
a Basal entropy (represented as a boxplot with whiskers where the interquartile range as extremities of the box, the median is the horizontal line in the box, and the whiskers as data extremes) recorded based on automated FC during chemostat experiments for the six biological systems investigated (values of entropy are computed from the population diversification profiles over the entire cultivation). Experiments have been done in duplicates (n biological replicate = 2) for each cellular system and cultivation conditions (chemostat or Segregostat) and exhibit a high reproducibility, see Supplementary Fig. 3). b Computation of the controllability (gain in entropy from chemostat to Segregated conditions represented as a boxplot with same structure than for plot A) for the six biological systems investigated. These average gains have been obtained by subtracting the mean value of H(t) recorded in Segregostat (considered as the controlled condition, i.e., cell population under environmental forcing) from the basal entropy. c Different diversification profiles, with different levels of controllability, can be observed based on the comparison of the H(t) profiles between non-controlled (chemostat) and controlled (Segregostat) conditions. d Single-cell traces of yeast Pglc3::GFP cells cultivated in a dMSCC device fluctuating between 1 and 0.1 mM of glucose (T1mM = 3 h; T0.1mM = 0.8 h). Between 10 and 40 cells have been tracked in four different cultivation chambers over two biological replicates (mean fluorescence is shown in bold). Pictures of a microcolony taken at regular time intervals are shown (Supplementary Movie 3). e Comparison of the mean fluorescence profile obtained in dMSCC with the ones obtained in classical MSCC at high (1 mM) and low (0.1 mM) glucose concentrations. The shaded region around the lines represents the standard deviation computed from the measurement (between 10 and 40 cells have been tracked in four different cultivation chambers over 4 biological replicates (n = 4) for each condition). f Mean values of the entropy over the whole microfluidics experiments run at different glucose concentrations and standard deviation across chambers (n microfluidic chambers = 4 observed over one experiment). Each mean value over a cultivation run in a chamber is represented as a dot, and the mean of all replicates with the standard deviation across them as a black dot with error bars. Source data are provided as a Source Data file.
