Fig. 4: Sensitivity and selectivity of the detection platforms.

a, b Calibration curves for miR-375-3p (a) and miR-141-3p (b) using asymmetric salt buffer (cis, 40 mM KCl; trans, 400 mM KCl). The concentrations of both molecular probes are 1 pM. As a comparison, data are shown using conventional single-molecule confocal microscopy and bulk fluorescence using molecular beacons at a concentration of 500 pM/100 nM. The dashed lines represent LODs at three times the standard deviation (3σ) of blank controls. The error bars represent 1 × σ obtained from 3 different nanopore measurements (n = 3). Data are presented as mean ± s.d. a.u., arbitrary units. c Sequences used for assessing the selectivity, consisting of the perfectly matched targets (let-7a, miR-141-3p), single-/double-nucleotide mutants (let-7f, miR-200a-3p), and scrambled sequences to Probe-let-7a and Probe-141. Variant bases are highlighted in red. d, e Box plots summarising the % synchronisation with these sequences. The black central line represents the median, the bottom and top edges mark the 25th and 75th percentiles, and the whiskers represent 1.5× interquartile range (IQR). Statistical significance was tested using a two-tailed Student’s t-test; ^****P < 0.0001. All the measurements were performed with at least 3 independent nanopores (n = 5 for let-7a/miR-141-3p; n = 4 for let-7f/miR-200a; n = 3 for scrambled sequences; n = 5 for blank controls).