Figure 3
Computational clustering of receptors. (a) A schematic of the mean-shift clustering algorithm used to group localizations detected by Exchange-PAINT into clusters based on local density. (b) In vitro clustering validation. Five types of DNA origami structures (a to e), each displaying 48 DNA-PAINT docking strands, were mixed and then imaged sequentially as previously described. (c) Super-resolved Exchange-PAINT image of the mixed origami sample on a glass surface. Imaging was performed using Cy3b-labelled imager strands at 10 nM, 15,000 frames per cycle, 10 Hz imaging rate. Washing: 1–2 minutes per cycle. Scale bar: 200 nm. (d) Localizations from the mixed origami sample in c were grouped into clusters (represented in different colours) of high local density as detected by mean-shift clustering. Cluster area is outlined in gray (dashed lines in the upper right corner are 3-fold higher magnification views of outline regions in c and d. Scale bars: 200 nm. (e) Clusters and their outlines in BT20 cells as detected by mean-shift clustering in the image shown in Fig. 2f. Scale bar: 500 nm. (f,g) Quantitative clustering analysis results after Exchange-PAINT imaging of 5 different receptors comparing unstimulated and EGF-stimulated BT20 cells.
