DNA Nanostructures Characterized via Dual Nanopore Resensing

TL;DR

Dual nanopore sensing with active feedback enables multiple translocations of a single DNA nanostructure, generating richer measurements (ΔG, dwell time, TOF, resensing timing) than single-pore approaches. The method uses machine learning on eight derived features from triple-scan events to classify closely related DNA origami seeds with accuracy above 0.83. A finite-element diffusion model links TOF to diffusion constants and particle lengths, providing quantitative size estimates and insight into inter-pore transport. This approach improves single-structure characterization in solution and offers a scalable, label-free tool for validating DNA nanostructure designs.

Abstract

DNA nanotechnology uses predictable interactions of nucleic acids to precisely engineer complex nanostructures. Characterizing these self-assembled structures at the single-structure level is crucial for validating their design and functionality. Nanopore sensing is a promising technique for this purpose as it is label-free, solution-based and high-throughput. Here, we present a device that incorporates dynamic feedback to control the translocation of DNA origami structures through and between two nanopores. We observe multiple translocations of the same molecule through the two distinct nanopores as well as measure its time-of-flight between the pores. We use machine learning classification methods in tandem with classical analysis of dwell-time/blockade distributions to analyze the complex multi-translocation events generated by different nanostructures. With this approach, we demonstrate the ability to distinguish DNA nanostructures of different lengths and/or small structural differences, all of which are difficult to detect using conventional, single-nanopore sensing. In addition, we develop a finite element diffusion model of the time-of-flight process and estimate nanostructure size. This work establishes the dual nanopore device as a powerful tool for DNA nanostructure characterization.

Figures (7)

• Figure 1: Dual-nanopore chip schematic. (a) A 3D schematic of the chip. (b) Zoomed-in view of the dual-nanopore structure. Two nanopores are positioned at the tips of the two "V"-shaped microchannels. Each nanopore thereby interfaces the common chamber above the membrane to a separate fluidic channel that can be independently addressed.
• Figure 2: resensing DNA nanostructures with active feedback control in a dual-pore device and representative signals. (a) Graphical depiction of a DNA nanostructure resensing cycle. A nanostructure is driven through pore 1 (left) under a positive voltage $V_1$ while the pore 2 voltage $V_2$ is held negative (i). Both voltages are then reversed so the same nanostructure translocates back through pore 1 (ii) and moves toward pore 2 (iii). Upon detection of nanostructure translocation through pore 2 (iv), the voltages are reversed again so the nanostructure translocates back through pore 2 (v). (b) The current traces ($I_1$, $I_2$) and voltages ($V_1$, $V_2$) during a resensing cycle. (c) Close-ups of the current signals of pore 1 sensing (i) and resensing (ii), definition of time-of-flight (TOF, iii), and pore 2 sensing (iv) and resensing (v).
• Figure 3: AFM images of DNA nanostructures used in the study with corresponding simplified and detailed schematics. Clockwise from the upper left are DNA nunchuck seeds, compact seed monomers, fringed-looped seeds, and looped seeds. The scale bars represent 400 nm in the main images and 40 nm in the insets. Nunchuck monomers, compact seeds and looped seeds are all comparable in length (77-80 nm), while fringed-looped seeds are about 12 nm shorter. In the simplified schematics, regions of double-stranded scaffold are yellow while regions of single-stranded scaffold are blue. Orange and green indicate the presence of adapter tiles and blocker strands, respectively. The dsDNA that links two compact seed monomers to form a nunchuck seed is dark blue. The detailed schematics positioned next to their respective simplified versions were generated using oxViewoxView.
• Figure 4: Density heatmap of conductance blockade versus dwell time for single-pore translocations. (a) Comparison between nunchuck and compact monomer seeds. (b) Comparison of compact seed monomers, looped seeds and fringed-looped cylinders. Darker shades correspond to a greater number of observations. On top and to the right are the corresponding violin plots (depicting area-normalized probability density functions derived from kernel density estimation, reflected about the binning axis) for each nanostructure. The box plots inside violin plots depict first quartile - 1.5 IQR (interquartile range), first quartile, median, third quartile and third quartile + 1.5 IQR. The distributions for the different seeds are stacked layer-by-layer. In (a) the nunchuck distribution is stacked on top and monomer below; in (b) the stacking order is compact, looped then fringed-looped seeds from bottom to top. Experiments were performed in 1 M LiCl buffer at 300 mV with a bandwidth of 30 kHz and a pore size of 37.5 nm.
• Figure 5: Machine learning classification. (a) Workflow for the classification of DNA origami seeds using multi-translocation events and Random Forest (RF) classifier. Confusion matrix for classification (b) of nunchuck seeds and their compact seed monomers and (c) of compact seed monomers, looped seeds, and fringed-looped cylinders. Values in the matrix indicate the number of testing multi-scan events. Experiments were performed using a dual-pore chip with a pore 1 diameter of 37.5 nm, a pore 2 diameter of 35 nm and a bandwidth of 30 kHz. (d) Comparison of three-seed classification accuracy using a single-scan (pore 1 sensing events), double-scan (pore 1 sensing-resensing events), and triple-scan (pore 1 sensing-resensing & pore 2 sensing events), at 10 kHz and 30 kHz bandwidth, respectively.