Fig. 5: Increased experimental complexity.

Results inferred from three experimental datasets where naturally no GT exists. a, d, g Experimental traces, offset in time and separated by dashed vertical lines, with donor and acceptor fluorescence intensity (green, red) and FRET efficiency (FRET, black), representative for the datasets used in (b, c), (e, f), (h, i), respectively, with n(traces): 134, 163, 118; and n(datapoints): 36,604, 37,067, 43,512; sampling rate = 33 Hz, time per datapoint = 30 ms. All three datasets were kindly provided by M. Schlierf. b, e, h FRET efficiency histograms and FRET efficiencies inferred by the analysis tools numbered as in (j). Sample sizes as in (a, d, g), respectively. The error bars represent standard deviations. For clarity, only the smallest reported model is shown for each analysis tool, up to a maximum of four FRET states. All inferred FRET efficiencies are shown in Supplementary Fig. 4, and all inferred results are provided in the Supplementary Tables 2–4 and in the Supplementary Datafiles. Purple arrow in (e): the error bar extends to 1.61. Teal arrow in (h): the error bar extends to −0.53. c, f, i Cumulative distribution functions (CDF) of the dwell-times simulated using the inferred kinetic models with two FRET states, obtained with the tools numbered as in (j). j Legend with all analysis tools. No results were reported by Edge finding. ‡ denotes results that were submitted after all other results were known. Source data are provided as a Source Data file for panels (b, c, e, f, h, and i).