Fig. 4: Resolving two constantly emitting fluorophores at distances down to 8 nm.
From: Diffraction minima resolve point scatterers at few hundredths of the wavelength
a, Measured distances d with respect to the expected distance (black line as specified by the manufacturer of the DNA nanoruler scaffold). Recording the photons originating near the illumination minimum suffices to separate simultaneously emitting fluorophores down to d = 8.4 nm within the uncertainty specified by the manufacturer (yellow shaded area). Circles indicate the medians of distances; error bars indicate twice the median absolute deviation from the median. Number of independent distance measurements per size of nanoruler (nm) are 8.4: 3,465, 10.2: 2,386, 14.3: 2,016, 20.1: 3,382, 25.0: 380, 31.8: 442. b, Exemplar localizations of two fluorophores attached to individual nanorulers with expected distances 8.4 nm (i), 10.4 nm (ii) and 14.8 nm (iii). The mean distance \(d\) as well as the average single-molecule localization precision (standard deviation of localizations) \({\sigma }_{x,y}\) is indicated. c, Corresponding histograms of the distance estimates. The standard deviation is indicated for each nanoruler. d, Box plot of obtained standard error of the mean (s.e.m.) for all nanorulers. The box extends from the lower to the upper quartile values of the data. Error bars mark the interval from the median (black line within box) to the last point within 1.5 × interquartile range (1.5 IQR). Outliers are shown as black dots. The median standard error of the mean is below 0.5 nm for all investigated systems and the 1.5 IQR is always <1 nm. Number of independent measurements are as in a. e, Box plot of the deviations from the expected distances. Number of independent measurements are as specified in a. The box extends from the lower to the upper quartile values of the data. Error bars mark the interval from the median (black line within box) to the last point within 1.5 IQR. Outliers are shown as black dots. f, Histogram of the distances obtained for the 8.4 nm fluorophore separation overlaid with the scaled probability density function (red line). The corresponding box plot from e is shown above. Comparing the total spread of the obtained deviations and the vastly different means of individual nanorulers within a batch indicates that the total spread may be partitioned into low measurement uncertainty and high production variability of the nanorulers (Extended Data Fig. 2 and Extended Data Table 1).
