Supplementary Figure 9: Classification criteria for cluster detection and sensitivity of STED image autocorrelation analysis to detect nanoclustering.

(a) Schematic of image autocorrelation curves of random (red) and different cluster scenarios (blue): cluster scenarios were classified as clearly detectable, when confidence intervals did not show overlap within the first two data points (left); as borderline, when confidence intervals overlapped within the first two data points, but mean values were outside the confidence intervals (center); as not detectable, when mean values were lying within the respective confidence intervals within the first two data points (right). All curves are plotted as mean ± SEM. The dashed line indicates that detectability was evaluated within the first two data points. (b) Image autocorrelation analysis was performed for different simulated clustering scenarios and assessed for the difference from simulated random molecular distributions; detectable difference (dark gray), borderline (light gray) and not detectable difference (white) as shown in a. Reference numbers indicate scenarios published in the literature. Simulations of nanoclusters with radii of 20, 40, 60, 80, 100 and 150 nm for 3, 5, 10, 15 and 20 clusters/µm². The fraction of molecules inside clusters was varied between 40% and 100%. Average molecular densities were adjusted to 75 molecules/µm² based on KT3-scFv-AS635P labeling of CD3ε (Fig. 5). Parameters of the simulated point spread function were based on experimental data. Numbers in boxes indicate the average number of molecules per cluster and the relative enrichment of molecules inside versus outside of clusters is indicated. The relative clustered area (right) was calculated from thresholded binary cluster maps. The figure is an extension of Fig. 6: panels for 20 nm and 60 nm are identical to Fig. 6