Fig. 4.

Simulation of microcolony morphogenesis and force prediction. a Measured aspect ratio of WT E. coli microcolonies grown in force microscopy experiments (red, N = 26) compared to simulations (black: N = 10, F_link = 4.25 pN). b When the microcolony expands in the plane (upper panel), both the elastic energy of the links (E_adh) and the repulsive energy of steric interactions (E_rep) increase. In contrast, when a bacterium goes into 3D and initiates second layer formation, it releases its repulsive energy and the microcolony saves a small area δA  in surface expansion . Yet, the bacterium must pay a cost in order to deform the gel and to extend its elastic links in the vertical direction. c Scaling between the size of the microcolony at second layer formation (N_2D/3D) and the force in adhesion foci (F_foci) for an agarose gel (15 kPa, black line). Thanks to the relation between N_2D/3D and F_foci (see Eq. (20) in Methods), we inferred the force at adhesion foci F_foci in glass–agarose experiments from the number of bacteria at the onset of the second layer (Fig. 1e). In the inset, comparison between the theory for a soft PAA gel (4 kPa, red dashed line) and experimental values reported in Fig. 3d (dark gray dots). d Comparison of colony shape between experimental data of WT E. coli microcolonies (red) in glass–agarose experiments (Fig. 1a) and simulations (polar, green; uniform, gray). e Comparison of the orientational order in microcolonies between experimental data on WT E.coli (red) and simulation (polar, green; uniform, gray). In d and e, the simulations are defined as polar when adhesive links form only at poles, and uniform when they form homogeneously along the bacterium