Fig 2: Calibration of a population-average model AModel calibration with median SMAD4 translocation dynamics of cells stimulated with different TGFß concentrations. Solid lines represent model fit, filled circles measured data. Experimental data points correspond to Appendix Fig S2J.B, CDynamics of total cellular TGFß receptor protein levels upon stimulation with 100 pM TGFß. Dynamics of TGFßR1 (B) and TGFßR2 (C) are shown as fold change relative to unstimulated cells. Solid red lines represent model fit and circles data measured by Western blot analysis. For each observation point, data spread from triplicates is shown. Gray area indicates measurement noise estimated by the fitting algorithm.DRestimulation with 2.5 pM TGFß. Stimulation was performed once at the beginning of the experiment (left), at the beginning and at 3 h (middle), or at the beginning and at 6 h (right) to replenish the ligand pool to its initial concentration. Solid lines represent model fits, and circles measured data points. Gray area indicates measurement noise estimated by the fitting algorithm.E, FModel calibration with median SMAD2 and SMAD4 translocation dynamics upon receptor inhibition. Cells were stimulated with 100 pM TGFß, TGFßRI activity was inhibited using the small-molecule receptor inhibitor SB431542 at the indicated time points (dashed vertical lines). Nuc/cyt SMAD2 (E) and SMAD4 (F) ratios were considered. Solid lines represent model fits, and circles or asterisks measured data. Experimental data points correspond to Fig 1G and Appendix Fig S2K. Gray area indicates measurement noise estimated by the fitting algorithm.GSmad7 mRNA induction upon 25 pM and 100 pM TGFb stimulation. Solid lines represent model fits, and circles measured fold changes relative to unstimulated cells. Experimental data points correspond to Fig EV2A. Gray area indicates measurement noise estimated by the fitting algorithm.HModel prediction of restimulation with 100 pM TGFß at 3 h (dashed vertical line). Solid line represents model prediction, and asterisks (*) measured data.

Fig 3: PA200 is a negative regulator of myofibroblast differentiation. (a) mRNA expression analysis of PA200 (PSME4), cyclin D1 (CCND1), TGF-ß1 (TGFB1) and aSMA (ACTA2) in phLF after 24, 48 and 72 h of PA200 silencing. HPRT served as housekeeping gene and expression was normalized to time-matching controls (one-sample t-test, phLF from n = 3 organ donors). (b) Protein expression analysis of PA200, myofibroblast marker a-smooth muscle actin (aSMA) and proliferation markers cyclin D1 (CCND1) and proliferating cell nuclear antigen (PCNA) in phLF transfected with PA200 or control siRNAs for 72 h. Bar diagram shows densitometric analysis of Western blots with normalization of obtained signals to control siRNA-transfected phLF of the same donor (one sample t-test, phLF from n = 4 different organ donors). (c) Cell count of phLF analyzed 72 h after PA200 silencing and normalized to mean of control siRNA-transfected cells (Mann-Whitney U test, phLF from n = 4 different organ donors with three technical replicates per condition). (d) Analysis of PA200 protein levels in phLF upon PA200 overexpression for 24 h (one-sample t-test, phLF from n = 3 different organ donors). (e) Analysis of cell count in phLF under the same conditions used in (d) (one-sample t-test, phLF from n = 3 different organ donors). (f) mRNA expression analysis of Psme4 (PA200), and myofibroblast markers Acta2 (aSMA) and Col1a1 in wildtype and PA200-/- primary mouse lung fibroblasts. Hprt served as house keeping gene (Mann-Whitney U test, pmLF from n = 4 animals per group). Full-length immunoblots are presented in Supplementary Fig. S5.

Fig 4: Classification of homogeneous populations. (a) Representative DHM images for two cell state transitions: EMT (top, NuMuG cells treated with TGFB) and DNA damage (bottom,? primary human melanocytes treated with Doxorubicin). (b) Two split representative feature correlation matrices for mammary epithelial cells undergoing EMT (bottom left) and human melanoma cells undergoing growth arrest (top right). Areas of conserved (purple) or non-conserved (green) correlations are highlighted. (c) Minimal correlation matrix from thirty-five experiments, each containing ~1,000–10,000 cells. (d) Representative DHM images of indicated treatments. Zoomed insets show similarity of individual cells within populations. (scale bars: 100microns (blue), 10microns (green)). (e) Strategy for DHC-based classification using cells of verified state. (f) Representative images of cells designated as M-phase, pre-apoptotic, or growth arrested. Cell fates verified after the analyzed 24-hour time-point. (g) Distribution of area and thickness for each cell state. (h) 2D and 3D scatter plots of feature distribution for each cell state. (i) Each plane from three-dimensional LDA space (Fig. S5) derived using twenty-six features, demonstrating clusters of pre-apoptotic (yellow), growth-arrested (red), non-treated (green) or M-phase (blue) cells. (j) Distribution of percent accuracy of cell classification across all experiments using single, double or triple feature sets versus machine-learning based phenotypic profiling. Plots g-i used 470 pre-apoptotic, 195 growth arrested, 66 M-phase, and 1527 non-treated cells.

Fig 5: Activation of TGF-ß1-Smad signaling by MAF up-regulated ß/?-crystallin expression.a Elevated expression of TGF-ß1 and p-Smad2/3 in primary human LECs with MAF overexpression (n = 3, MAF p = 0.002, TGF-ß1 p = 0.0002, p-Smad2/3 p = 0.002). Right, the band density in Western blotting was normalized to loading control as a ratio for statistical analysis. b Elevated secretion of TGF-ß1 into cell culture media of primary human LECs with MAF overexpression (n = 4, p = 0.043). c, d Changes of Tgfb1 expression in primary mouse LECs in accordance with Maf changes as detected by qPCR (both n = 6, p = 4.36e-5 and p = 1.16e-7) and Western blotting (n = 3, Maf OE: MAF p = 0.001, TGF-ß1 p = 2.30e-5; Maf KD: MAF p = 0.003, TGF-ß1 p = 0.010). Right, the band density in Western blotting was normalized to loading control as a ratio for statistical analysis. e Elevated transcriptional activity of TGFB1 gene with co-transfection of MAF plasmid (p = 0.0007). f, g Changes of ß/?-crystallin gene expression detected by qPCR (n = 6, Crybb1: control vs. Maf OE and Maf OE vs. Maf OE + TGF-ßR1/2 inhibitor p = 3.30e-7 and p = 6.23e-8; Crygd: p = 0.001 and p = 0.027, Cryba2: p = 1.7e-5 and p = 0.001; Cryba4: p = 0.005 and p = 0.011; Cryba1: p = 0.007 and p = 0.028) and Western blotting (n = 3, CRYBB1: p = 0.008 and p = 0.019, CRYGD: p = 9.2e-5 and p = 6.2e-5, CRYBA2: p = 0.0005 and p = 6.4e-5, CRYBA4: p = 0.003 and p = 0.001, CRYBA1: p = 0.015 and p = 3.5e-5) with Maf overexpression ± TGF-ßR1/2 inhibitor treatment (10 µM for 24 h) in primary mouse LECs. Right, the band density in Western blotting was normalized to loading control as a ratio for statistical analysis. h No significant change of MAF detected by qPCR (both n = 4, p = 0.586 and p = 0.511) in primary human or mouse LECs after TGF-ß1 treatment (5 ng/ml, 24 h). n = biological replicates. Results are expressed as mean ± SD. Level of significance was detected using two-sided Student’s t test (a–e, and h) and one-way ANOVA plus Tukey’s multiple comparisons test (f and g). ****p < 0.0001, ***p < 0.001, **p < 0.01, *p < 0.05 and ns represents no significant difference. Source data are provided as a Source Data file.