Fig. 6. Comparison of model to experimental data.
a, b Estimates were obtained from data for the parameters normalized apical length (la) (a) and size regulation stringency (k) (b) as described in “Methods” section. The parameters p and lb were free parameters and tuned to agree with data, with the constraint that p for the Cldn-qKO condition was ten-fold smaller than for the other two conditions. c, d Normalized lumen cross-sectional area (c) and normalized mean basal length (d) and as a function of size for WT, Rab11a-GFP OE, and Cldn-qKO spheroids. Model predictions (lines) show good quantitative agreement with experimental data (points). Welch’s two-sided t-test with Coarsened exact matching (CEM) for cross-sectional lumen area for WT vs. Rab11a-GFP OE p = 0.7, WT vs. Cldn-qKO p = 5 × 10−4. Welch’s two-sided t-test with CEM for basal length for WT vs. Rab11a-GFP OE p = 0.9, WT vs. Cldn-qKO p = 9 × 10−6. Welch’s two-sided t-test with CEM for simulation cross-sectional area with la = 0.6 vs. 0.8, p = 0.06; la = 0.6 vs. 1.0, p = 8 × 10−4. e Plot of model predictions of lumen solidity as a function of number of cells. Welch’s two-sided t-test with CEM for simulation cross-sectional area with la = 0.6 vs. 0.8, p = 0.50; la = 0.6 vs. 1.0, p = 0.38. f Schematic depicting pressure-dependent and pressure-independent lumenogenesis. For plots a, c, and d n = 51 WT, 51 Rab11a-GFP OE, and 25 Cldn-qKO conditions. Error bars on lines indicate standard deviation for simulation results, center of error bars on lines indicate mean of simulation results, for n = 25 simulations for plots c, d, and e. Source data are provided in the Source Data file.