Advances of nanopore sensors toward virus detection and diagnostic applications

Abstract

With the advantages of ultra-sensitivity and high throughput, nanopore technology has now evolved into a versatile tool for a wide range of practical applications, including genomic sequencing, proteomic analysis, and detection of various infectious and noninfectious diseases using biomarkers. Especially for infectious diseases, the rapid diagnosis of pathogenic microorganisms is a critical prerequisite for pandemic control and treatment. It is well known that the whole-genome sequences of some pandemic viruses have been accomplished to provide a high-resolution view of pathogen surveillance. This article reviews the progress of nanopore sensors towards virus detection and clinical applications, focusing on innovative strategies aimed at enhancing the detection efficiency. Intrinsically, the nanopore allows the single-molecule counting of viruses in nanofluidic channels. Some nucleic acid and protein components of the viruses are also potential target candidates for virus detection. Meanwhile, a variety of molecular probes involving aptamers, nucleic acids, peptides and nanoparticles have been designed to improve the detection sensitivity of target viruses. The stochastic sensing mode of nanopores further simplifies the conventional testing process, focusing on the rapid and qualitative identification of multiplex viruses, making it more feasible for portable, point-of-care diagnostics.

---

1. Introduction

Viral diseases pose severe threats to life and health, ranging from mild illnesses to deadly outbreaks, which will cause unprecedented global impact on life safety, social economy and political stability.^1–3 Especially, since the severe acute respiratory syndrome-coronavirus 2 (SARS-CoV-2) pandemic, viral detection technologies have become increasingly important, making them a critical part of global public health control systems.^4–6 Earlier, viral detection mainly depended on observing the symptoms caused by the pathogens and indirectly determining the presence of viruses by some basic testing means, such as inoculation and infectivity testing, microscopic examination, serological testing, and cell culture techniques.^7–9 Along with technical progress, the conventional diagnostic methods used for virus detection include polymerase chain reaction (PCR) test, enzyme-linked immunosorbent assay (ELISA), and gene sequencing.^10,11 However, these techniques are limited by the need for expensive specialized equipment, complex procedures, and long processing times; furthermore, they still cannot meet the needs of the extensive surveillance required for the high mutation rate and genetic diversity of viruses in clinical trials. In this situation, a facile method for the rapid and precise assay of pathogenic viruses, which can be more widely used, is urgently required.

Nanopore technology is becoming a robust tool for virus detection due to its unique single-molecule detection capability, convenient operation process, and potential portability.^12–21 It is inspired by the working principles of early Coulter counters, and its sensing method involves detecting the individual transport events of analytes by single-channel current recording, which enables it to go beyond ensemble-averaged measurements. With the improvements in micro- and nano-fabrication, the nanopore has achieved a more refined control over pore dimensions and expanded its application scope, enabling single-molecular detection at a much finer level.²² It has now evolved into a high through-put and label-free sequencing platform with its long-read capability.^23–25 The whole-genome sequencing of some pandemic viruses, such as the coronavirus,^26,27 Ebola virus,^28,29 and monkeypox virus, has been accomplished to provide a high-resolution view of pathogen evolution.^30,31 Nanopores allow single-particle counting of viruses in biological specimens, and some viral species with different shapes and surface features can be recognized based on the current blockage of nanopores.^32–38 These powerful viral detection capabilities have been given attention and widely promoted in some important works.^19–21 However, in this emerging and rapidly developing field, more research efforts are focusing on the in-depth exploration of nanopores and improving the detection accuracy to meet the demands of clinical diagnosis in various scenarios.

As viruses are principally composed of internal genetic material (nucleic acids) and surface protein shells, these nucleic acid fragments and protein components are potential target candidates for monitoring viral mutants and immune responses due to their portability and real-time response characteristics in on-site rapid diagnostics.^39–44 To improve the detection sensitivity of the target viruses, multifarious molecular probes and design strategies, such as aptamers, DNA/peptide probes, nanoparticles, and specific modification of the pore surface, have been employed to simultaneously recognize multiple antigen proteins and gene sequences of target viruses.^45–52 In the process of recognition, the resulting products can be recorded by the nanopore sensors integrated with other advanced technologies like microfluidics, nanomaterials, electrochemical systems and optical sensing.^53–58 The integration platforms are committed to handling the detailed problems of nanopore sensing existing in sample preparation, fluid transport, and signal acquisition.^53–57 The growing application of machine learning algorithms to nanopore signal processing is helping to promote the digital pathology platform of virus databases. Therefore, the latest strategies and potential applications of nanopore technologies provide rapid and valid solutions for the accurate diagnosis of virus infections and outbreak control.

2. Nanopore sensors

The basic principle of nanopore sensing technology involves the high-precision detection of individual molecules through nanoscale pores. A typical nanopore device employs a dual-chamber microfluidic architecture, where an insulating membrane (e.g., lipid bilayer or solid-state thin film) embedded with a nanoscale pore separates the system into two independent compartments of the cis chamber and trans chamber. Both chambers are filled with conductive electrolyte solution (typically potassium chloride solution), with the inserted Ag/AgCl electrodes for the applied voltages. Under certain voltages, ions in the electrolyte solution move directionally through the pore, forming a steady baseline current, while single molecules or particles are driven into the nanochannel under the combined effects of electroosmotic and electrophoretic forces.^59–61 The passing molecule/particle temporarily displaces an equivalent volume of the electrolyte solution, which results in a transient fluctuation of conductivity changes within the nanopore, termed a detectable current pulse signal. Generally, these pulse signals are caused by the combined effects of volume exclusion and surface charge of the molecules passing through the pores. These current variations are recorded in real time by a transimpedance amplifier system, enabling the identification and characterization of individual molecules at the single-particle level.⁶²

Nowadays, nanopore sensors are primarily categorized into two major types, namely, biological and solid-state nanopores. Biological nanopores are composed of membrane proteins inserted in the lipid membranes to form highly confined spaces, enabling high-precision molecular recognition.⁶³ Solid-state nanopores are constructed from synthetic materials, including silicon nitride films, glass nanotubes, two-dimensional materials, and so on. These ultrathin films are drilled at the nanometer scale by various micro-nano fabrication techniques, like electron beam lithography, electron/ion beam milling, and the dielectric breakdown technique.²² With excellent mechanical stability and controllable pore dimensions, solid nanopores allow the translocation of molecules and particles of different sizes, greatly expanding the range of nanopore applications.

So far, a variety of nanopore platforms have been significantly developed for label-free single-molecule sensing and sequencing applications. In particular, the emergence of nanopore sequencing is a competitive technology, with the long-read, label-free and high-throughput advantages, overwhelming the read-length limitations of traditional gene sequencing. After continuous product updates and iterations, nanopore-based DNA sequencing devices have been developed for commercial use. The available sequencers from Oxford Nanopore Technologies (ONT) have been widely commercialized in genomics and clinical diagnosis on a large scale.⁶⁴ Besides DNA sequencing, nanopores are also emerging as a new application for protein sequencing, drug screening, and pathogen detection.⁶⁵ Viral identification using nanopores is still in its early stages, and more in-depth research is expected to improve the accuracy and throughput of pathogen detection.

3. Nanopore sequencing for viral detection

Genomic sequencing is considered the most direct and effective way to identify the causative agents of infectious diseases, implying viral pathogenicity, evolution, and transmission. Recently, the nanopore platform has been applied to directly sequence the DNA/RNA of bacterial and viral samples, for the studies of genomic surveillance and epigenetic changes in transcriptomics and metagenomics.⁶⁶ In particular, the complete genome sequencing of SARS-CoV-2 by ONT provides insights into the virus variation, with important implications for therapeutic and preventive measures.^18,26 The long-read nanopore sequencing in real time is rapid, cheap and portable with flexible scalability, but the concerns regarding its sequencing accuracy over traditional sequencing technology remain. To assess the validity of nanopore sequencing, taking SARS-CoV-2 genomics as an example, Bull and colleagues have systematically evaluated the analytical performance of Oxford nanopore and Illumina sequencing platforms, as shown in Fig. 1(A). Despite their distinct error profiles across the SARS-CoV-2 genome, the sequencing accuracy reaches a high degree of consensus for both nanopore and Illumina sequencing platforms. The variant frequencies for single-nucleotide variants (SNVs) identified by both technologies are correlated (R² = 0.69), and SNVs at sub-consensus variant frequencies can be detected with high sensitivity and good precision. ONT sequencing also reveals the diversity of genomic variation within SARS-CoV-2 specimens, although it is not sufficient to discern short indels and variants at low read-count frequencies exactly. The valid comparative performance makes it better for viral detection and genomic analysis, complemented with conventional sequencing technologies. Certainly, nanopore sequencing requires further breakthroughs in engineering nanopore chips, pathogen genetic libraries and bioinformatics software, for application in the diagnosis and control of epidemics in the future.

Fig. 1 (A) Analytical performance of whole-genome sequencing of SARS-CoV-2 based on both Oxford nanopore and Illumina platforms, and the correlation of variant frequencies observed for SNV candidates detected at sub-consensus frequencies (20–80%).¹⁸ (B) Target amplification of nanopore sequencing and the real-time reverse transcription-polymerase chain reaction method.¹² Reproduced from ref. 18 with permission from Springer Nature Limited, copyright 2020. Reproduced from ref. 12 with permission from WILEY-VCH Verlag GmbH & Co., copyright 2020.

More respiratory viruses have been distinguished simultaneously by nanopore-targeted sequencing.¹² Viral nucleic acids encompass genomic DNA, RNA, and the transcriptome. The diversity in viral genome types and structures necessitates distinct library preparation and sequencing approaches. As shown in Fig. 1(B), the ORF1ab and virulence factor-encoding regions are designed as targets and amplified by a set of specific primers to develop the SARS-CoV-2 primer panel. The multiplex amplified targets are sequenced to confirm the virus species by gene mapping, coverage, and read number. The nanopore target sequencing is capable of distinguishing different types of respiratory viral infections simultaneously, providing rich supplemental data for the epidemiological survey of viruses. For better broad-spectrum pathogen identification, metagenomic sequencing with spiked primer enrichment is performed on both Illumina and nanopore sequencing platforms.¹⁴ It can enrich target RNA sequences while preserving metagenomic sensitivity for all pathogens, thus promising a broad application prospect for clinical diagnosis and outbreak surveillance.

4. Single-particle counting of viruses by nanopores

4.1. Single-particle characterization of viruses

By virtue of the unique transport properties, nanopore sensors can directly discriminate individual viral particles in the single-molecular counting form, without requiring genomic extraction.^{34,37,38,61,67–69} As virus particles typically range in size from tens to hundreds of nanometers, solid-state nanopores with tunable geometry are favorable platforms for the detection of virus entities. For instance, Uram et al. early used a submicrometer glass pore to detect single chlorella virus particles and evaluate the binding affinity of the virus for the antibody.⁶⁷ Zhou et al. employed single track-etched conical nanopores formed in poly(ethyleneterephthalate) (PET) membranes to distinguish hepatitis B virus (HBV) capsids.⁷⁰ More virus particles with different geometries and surface charges have been identified and evaluated by nanopores, like spherical, polyhedral and filamentous virus particles. Typically, McMullen and colleagues studied the voltage-driven dynamics of filamentous virus fd through silicon nitride nanopores.³² As shown in Fig. 2(A), a stiff fd virus must linearly translocate through the pore. The electric field distribution can align and capture fd viruses approaching the nanopore, even exhibiting some collision attempts. The translocation dynamics reveals voltage-independent mobility and follows first-passage time distribution, which could provide theoretical guidance for virus capture and translocation through nanochannels.

Fig. 2 (A) Studies of filamentous (fd) viruses through nanopores and the typical electrical signals from fd interacting with nanopores.³² Two distinct event populations can be seen in the scatter plots of fd translocation at 120 mV (right panel). (B) Schematic of virus detection in a peptide-decorated Au nanopore and magnified views of the resistive pulses for single-virus detection.³⁷ Ionic current curves acquired by influenza A(H1N1) in Si₃N₄ (orange) and in P₂ nanopores (red) (right panel). (C) Infectious HAdV detection using aptamer-functionalized nanopore sensors.³⁶ Normalized rectification efficiencies (f_recnorm) versus virus concentration are shown in nanopores without aptamer (black), and nanopores modified with aptamers for noninfectious virus (green) and infectious virus (purple). The f_recnorm is the ratio of each f_rec from the samples with and without viruses. Reproduced from ref. 32 with permission from Springer Nature Limited, copyright 2014. Reproduced from ref. 37 with permission from the American Chemical Society, copyright 2018. Reproduced from ref. 36 with permission from the American Association for the Advancement of Science, copyright 2021.

4.2. Specific detection of viruses by functionalized nanopores

The nanopore has demonstrated ultimate sensitivity to detect diverse viral species depending on the physical properties of bioparticles in a label-free fashion, and the next issue is the discrimination of similar virus particles, as with different viral subtypes. For example, Kawai's group employed silicon nitride nanopores to discern three types of influenza viruses A (H1N1, H3N2) and B, with common spherical shapes and sizes but different surface charge densities.³⁸ The subtle differences of viral subtypes are often attributed to their surface proteins. They designed biorecognition solid-state nanopores for selective single-virus identification.³⁷ As shown in Fig. 2(B), the hemagglutinin antibody-mimicking oligopeptides are decorated in the pore wall surface as recognition probes to capture influenza A. The peptide-modified nanopores enable the discrimination of influenza A(H1N1) and B from the translocation dynamics, as H1N1 is prone to dwelling in the pore longer due to the enhanced specific binding between viruses and nanopores. The designability of synthetic peptide probes can offer more selective sites for immunological recognition, thereby facilitating single-virus screening by nanopores.

According to the functional surface of nanopores, Peinetti et al. designed a set of highly selective DNA aptamers to modify solid-state nanopores for detecting infectious viruses in real samples, without sample pretreatment.³⁶ As shown in Fig. 2(C), the DNA aptamers of target viruses are screened using the systematic evolution of ligands by exponential enrichment (SELEX), which is performed for a mixture of multiple viruses, minimizing the false positive and false negative errors. These DNA aptamers of specific sequences can form certain stereoscopic structures for better recognition of target viruses with high affinity. For conical nanochannels without aptamers (black), and with aptamers for noninfectious viruses (green) and infectious HAdV (purple), an asymmetric electrical current–voltage characteristic is measured, defined as the ion current rectification efficiency (f_rec). The normalized rectification efficiency (f_recnorm) decreases rapidly in response to changes in viral concentration. The infectious HAdV is quantified with a limit down to 1 pfu ml⁻¹. The nanopore sensors functionalized with DNA aptamers have demonstrated the direct detection of adenovirus (HAdV) and coronavirus (SARS-CoV-2) in complex environments, with enhanced precision and lower detection limits. This direct and ultra-sensitive detection offers good potential for the rapid diagnosis of viral pathogens in real scenarios.

4.3. Machine learning-assisted digital diagnosis of viruses with nanopores

Essentially, these current blockade signals are mainly caused by the volume exclusion effect of the bioparticles in nanopores, and the magnitude and number of these pulses are proportional to the volume and number of the particles, enabling their counting and size analysis. Hence, the similarity and variability of particles are challenges to direct detection, regardless of the viral species and subspecies. To process the signals obtained by viral specimens with similar characteristics, machine learning algorithms have been used to discriminate subtle differences in noisy signals without specific probe recognition.^33,34,68 Kawai and colleagues designed a digital pathology platform for multiplex single-particle detection, based on their studies of machine learning methods, as shown in Fig. 3(A).³⁴ Here, virus identification evolves into a topic of classification, and many models for classification, including Naive Bayes, SVM, random forest and rotation forest methods, have been used to extract the hidden features from the training data and make the classification decision.^34,38 In addition to conventional statistical analysis on peak amplitudes (I_p) and duration times (t_d) of current pulse signals, more features concerning the angle (θ), area (S), asymmetry (r), bluntness (β), and three kinds of inertia (I, I_wv, and I_w) have been extracted to reach a higher dimensional signal space. The multidimensional feature parameters described above have been randomly selected to form the classification criterion for a precise decision. Through machine learning-based signal analysis, F-measure scores are estimated for five viral species, including the respiratory syncytial virus (RSV), adenovirus, influenza viruses, and coronavirus, as shown in Fig. 3(A). Here, F-measure scores are defined as follows: F_meas = 2P_preP_rec/(P_pre + P_rec), where P_pre and P_rec are the precision and recall derived from the number of true-positive, false-positive, and false-negative cases. It is estimated to be 71.3% by only one current pulse, and is enhanced with the number of current signals. There is over 99% accuracy for five different virus species by only several dozen signals, indicating the powerful capability of signal recognition processes.³⁴

Fig. 3 (A) Typical current traces and identification results of various virions (RSV, coronavirus, influenza A and B, and adenovirus) using the SiNx nanopore (d_pore = 300 nm, L_pore = 50 nm) under biased voltage (V_b = +0.1 V). The feature parameters of the resistive pulse are used for viral discrimination via machine learning.³⁴ (B) Nanopore measurements and ionic current–time traces of cultured coronavirus; the confusion matrix and identification accuracy obtained by machine learning, corresponding to the cultured coronaviruses.³³ Reproduced from ref. 34 with permission from the American Chemical Society, copyright 2020. Reproduced from ref. 33 with permission from Springer Nature Limited, copyright 2021.

These experimental results show that the performance of machine learning methods has significant potential to improve the accuracy and throughput of viral infection diagnosis, replacing the costly and time-consuming process of genome extraction and amplification. Thus, Taniguchi et al. further developed a machine learning-assisted nanopore analysis platform with high speed and high precision, consisting of machine learning software, a portable current measuring instrument, and scalable semiconducting nanopore modules.³³ As shown in Fig. 3(B), four subtypes of coronaviruses similar in size, involving HCoV-229E, SARS-CoV, MERS-CoV, and SARS-CoV-2, have been identified in saliva specimens, with a sensitivity of 90% and specificity of 96% with a 5-minute measurement. The machine learning-assisted nanopore analysis significantly enhances the detection speed, further reduces the operational difficulty and cost, and is more suitable for immediate detection scenarios. As a portable virus diagnostic system, it will promote the digital pathology platform of virus databases, which is valuable for better responses to infectious diseases and outbreak control.

4.4. The physical properties of virus particles in nanopores

Besides the ability to rapidly detect different viral species, nanopores have also been developed to characterize the physical properties of virions at the single-particle level. In principle, the transient current pulse signals are induced by particles passing through the pore; hence, viral particles are measured one-by-one in real time, which can provide the detailed relationships between the particle shape and surface charge, biological activity and epidemics.^69,71–73 For example, Arjmandi et al. developed a universal and precise model to measure the surface zeta-potential of model particles and viruses by monitoring the dwell time of particles in a nanopore based on biased voltages.⁷² The zeta-potential was monitored for both the human immunodeficiency virus (HIV) and the Epstein–Barr virus (EBV), as shown in Fig. 4(A). Briefly, the zeta-potential of the nanoparticles was extracted by multiplying the slope of 1/t_Tr versus voltage curve by A = (ηl²)/(ε_rε₀), written as . Here, t_Tr is the measured translocation duration, V is the applied voltage, ε_rε₀ and η are the solution permittivity and viscosity inside the nanopore, and l is a fitting parameter that is included in a calibration measurement from a particle of known zeta-potential. This approach not only enables the detection of nanoscale particles with similar sizes but also allows for the direct determination of their surface charges, especially for the low electric potential of these viruses in conditions of low concentrations and small volumes. Likewise, the particle masses and sedimentations of viruses have also been measured in the liquid phase by nanopores.⁷³ The nanopore technology seems to be an alternative tool for the simple and precise detection and characterization of nanometer-sized particles and viruses dispersed in a liquid without any labeling, beyond the limit of high-resolution transmission electron microscopy and dynamic light scattering in practical applications.

Fig. 4 (A) Nanopore device and zeta-potential of the nanoparticles; HIV and EBV viruses measured by the nanopore device versus their zeta-potential measured by a commercial DLS system.⁷² (B) Identifying DNA lengths in single vectors by the ion blockade characteristics; the resistive pulses were observed more frequently for the vectors with shorter DNA than those with longer DNA.⁶⁹ (C) Zero-mode waveguide setup for virus translocation through nanopores and the translocation frequency as a function of pressure for HIV particles at different concentrations.⁷⁵ Reproduced from ref. 72 with permission from the American Chemical Society, copyright 2012. Reproduced from ref. 69 with permission from the American Chemical Society, copyright 2024. Reproduced from ref. 75 with permission from Springer Nature Limited, copyright 2024.

Viral particles are like nanoscale capsules that carry genetic materials in the protein cage of the capsid. The hollow and DNA-filled viral particles can be discriminated by nanopore sensing.^61,69,74 Recently, Tsutsui et al. used adeno-associated virus (AAV) vectors as gene delivery vehicles to nondestructively examine genomes inside the viral capsid at the single-particle level. As shown in Fig. 4(B), the volume of the adenovirus vector increases with the genome length, and the resistive pulses are observed more frequently for AAV vectors with shorter DNA than those with longer DNA. They further used the viscosity of salt water–organic mixtures to slow the translocation of viral vectors in pores, which allows the precise tracking of the dynamic motions of AAV vectors in nanopores.⁶¹ These findings make it possible to noninvasively screen full, empty, and intermediate viral vectors with their genetic content based on the transport properties of a nanofluidic channel, which can serve as a promising tool to inspect the quality control of the gene delivery system using vector products.

Viruses are highly infectious in view of their complex interactions with the environment. The interactions among particles tend to aggregate in confined environments. Montel groups have unveiled a soft jamming phenomenon associated with virus transport in the ionic flow of nanopores.⁷⁵ As shown in Fig. 4(C), the translocation frequency of HIV is measured as a function of pressure for different virus concentrations. At high pressure, the increase in concentration led to a decrease in translocation frequency, which is associated with the clog phenomenon due to the interactions of viral particles with themselves and the pore surface. By proposing a quantitative model of virus jamming, the results provide valuable insight for understanding crowding behavior in confined spaces of viral particles and their interactions in transport across the membrane. As soft particles, viruses have unique morphologies and surface properties, as well as complex interactions with their environment, which will affect their transport characteristics through ion channels. Uncovering more transport dynamics of viruses through nanopores is crucial for resolving existing issues, such as clogging and deformation, thereby quantifying particle translocation.

5. Single-molecular detection of viral biomarkers by nanopores

5.1. Detection of nucleic acid biomarkers

As virus particles are composed of internal genetic materials and surface protein capsules, these nucleic acids and proteins are potential target candidates in the on-site rapid diagnostics and monitoring of viral mutants and specific targets due to their portability and real-time response characteristics.^47,52 Typically, Oh et al. used α-hemolysin to detect the conserved Influenza A virus (IAV) RNA promoter at the single-molecule level.⁵² As shown in Fig. 5(A), the IAV genome contains negative-sense RNA segments with some conserved sequences. These conserved sequences form a duplex panhandle structure that is involved in regulating viral transcription and replication, referred to as the RNA promoter, which is an important testing target of influenza viruses. Hence, they designed unique DNA probes that can complementarily hybridize with the RNA promoter with high specificity. There are two kinds of representative signal events induced from the RNA/DNA complex in nanopores, thus achieving the ultrasensitive and rapid distinction of the IAV promoter from nonspecific molecules.

Fig. 5 (A) Nanopore-based detection of free IAVp RNA and the representative nanopore events: the events of type I and for IAV promoters or DNA probes and the events of type III and IV for the complex.⁵² (B) Diagram of the selective solid-state nanopore assay and current trace for the translocation of biotinylated duplex nucleic acids alone, monovalent streptavidin protein alone, and duplex nucleic acids bound to monovalent streptavidin.⁴⁵ (C) Rapid detection of coronavirus SARS-CoV-2 by RT-LAMP-coupled solid-state nanopores and the nanopore event rate as a function of RT-LAMP reaction time.⁴⁴ Reproduced from ref. 52 with permission from the American Chemical Society, copyright 2019. Reproduced from ref. 45 with permission from the American Chemical Society, copyright 2021. Reproduced from ref. 44 with permission from Elsevier, copyright 2022.

Similarly, solid-state nanopores have been employed for the direct detection of viral genomic sequences.⁴⁵ The Hall group developed a two-component assay involving a short nucleic acid target with a single biotin tag through selective solid nanopores, as shown in Fig. 5(B). After the extraction of sequence motifs, the biotinylated DNA oligonucleotide is driven into the nanopores and bound to the monovalent streptavidin decorated in pores to generate a translocation event due to steric hindrance, yielding the sensitive detection of target sequences. Based on this concept, they demonstrated human pathogen detection of RNA viruses by identifying the conserved target sequences within human immunodeficiency virus (HIV-1B). The selective nanopores can be used to determine low abundance of sequence motifs under a mixed background with non-target oligonucleotides in fluid phases.

The conventional viral genomic assay often involves gene amplification based on PCR to meet the sample requirements.⁷⁶ It requires complex thermocycling, which is time-consuming with a limited throughput. Alternatively, the Guan group reported a reverse transcription loop-mediated isothermal amplification (RT-LAMP) coupled glass nanopore digital counting method for the rapid detection of SARS-CoV-2.⁴⁴ As shown in Fig. 5(C), the purified SARS-CoV-2 viral RNA can be rapidly accumulated by RT-LAMP at a constant temperature under simple conditions. The resulting products are then measured by glass nanopores. Some representative current traces are recorded corresponding to the negative control groups and the positive targets. The event rate of nanopore readout is a function of RT-LAMP reaction time, and a threshold is set as the criterion for positive cases, which is much higher than the background event rate in the negative control, minimizing the interference of the false-positive events. These RT-LAMP-coupled nanopore sensors have been demonstrated for both saliva samples and nasopharyngeal swab samples. Focusing on the viral detection, they also developed CRISPR-Cas12a-aided glass nanopores for more specific discrimination of SARS-CoV-2, HIV-1, monkeypox virus and so on.^41–43 Moreover, the nanopores combined with CRISPR technology enable the real-time monitoring of the gene editing process, providing possibilities for gene therapy. These highly specific sensing strategies simultaneously identify different DNA targets in complicated environments, revealing a unique way toward the rapid detection of nucleic acid analytes.

5.2. Detection of protein biomarkers

Proteins are also used as valuable biomarkers for virus diagnosis. For example, Galenkamp et al. designed a Cytolysin A nanopore (ClyA) lodged with proteins to detect glucose and asparagine.⁷⁷ Kwaw et al. used a ClyA platform to probe the enzymatic activity of neuraminidase (NA), enabling the rapid and sensitive diagnosis of the influenza A virus (IAV).⁷⁸ As the major capsid protein of IAV, NA is not only a useful biomarker for IAV diagnostics, but also a potential drug target for clinical treatments. During IAV infection, NA, as a glycoside hydrolase, cleaves the terminal sialic acids of glycan from the host cell surface to release viral progeny. As shown in Fig. 6(A), the cleavage activity of NA is quantitatively measured by ClyA nanopores, in which galactose substrates are enzymatically catalyzed and the released galactoses are real-time monitored according to the conformational transformation (open and closed) of the trapped glucose binding proteins. Moreover, the susceptibility and resistance testing of a set of small-molecule antiviral drugs was explored by assessing the inhibition of NA activity in nanopores. Similarly, Zhou et al. also developed the multiplexed detection of viral proteases for SARS-CoV-2, according to the cleaved peptide substrates during protease digestion.⁷⁹ Thus, the nanopore-based protease detection strategy may have potential application prospects in infectious disease diagnosis and drug screening.

Fig. 6 (A) Illustration of sialyl-galactose substrate cleavage by IAV NA, and the representative current trace for the proteins in the ligand-free (open) and ligand-bound (closed) states. The fractional times of L1 (f_L1) versus concentrations of the SG substrate fitted to a Hill function with the coefficient set to 1.⁷⁸ (B) The translocation of the NCp7 protein bound to RNA of the SL3 stem-loop recognition, and the representative electrical traces for the effect of N-ethylmaleimide on the formation of the NCp7 SL3 RNA complex.⁴⁷ Reproduced from ref. 78 with permission from the American Chemical Society, copyright 2020. Reproduced from ref. 47 with permission from the American Chemical Society, copyright 2013.

The viral capsid proteins play a key role in viruses invading cells and replicating themselves. For instance, Niedzwiecki et al. explored the single-molecule detection of the nucleocapsid protein 7 (NCp7) by solid-state nanopores.⁴⁷ NCp7 is a nucleic acid chaperone responsible for the reverse transcription of viral RNA and genome packaging of the human immunodeficiency virus 1(HIV-1) life cycle.⁸⁰ Based on specific interactions between NCp7 and aptamers of stem-loop 3 (SL3) in the packaging domain of the retroviral RNA genome, the binding affinities of NCp7 with SL3 and RNA aptamers have been systematically measured in real time by solid-state nanopores, as shown in Fig. 6(B), and the inhibition effect of N-ethylmaleimide has been verified by suppressing the formation of the NCp7-SL3 complex. Meanwhile, other surface enveloped proteins have been explored for virus diagnosis and drug screening. Cai et al. used the sandwich formation of the target and antibody to achieve single-molecule detection of monkeypox virus (MPXV) A29 protein directly in biofluids mixed with vaccinia virus A27 protein and varicella zoster virus proteins.^48,81 Zhao et al. discriminated the hepatitis B virus (HBV) and hepatitis D virus (HDV) by relying on the surface antigen–antibody reaction in vitro assay.³⁹ The nanopore-based biomarker detection will contribute to minimizing virus transmission.

Nanopore detection has shown potential in sensing subtle protein and enzyme conformation changes in aqueous environments. However, the protein biomarkers are often in low abundance in body fluids, crowded with numerous biomolecules and metabolites, which poses challenges for clinical diagnosis. With the advancement of nanopore technology, protein detection has been advancing towards the direct sampling of bodily fluids.^77,82–85 As mentioned above, the ClyA nanopore can monitor the substrate recognition of proteins lodged inside the pore to quantify the concentrations of glucose and asparagine from blood, sweat, and other bodily fluid samples.⁷⁷ Recently, Maglia groups have also reported that nanopores engineered with a polypeptide mesh as a selective entropic gate can detect proteins in complex biological samples.⁸² These nanopores can directly detect trace amounts of biological samples without sample preparation, making them ideal for large-scale virus detection.

5.3. Multiple detections of molecular biomarkers

With the development of molecular diagnostics for infectious diseases, multiple biomarkers, including nucleic acids/proteins, have been simultaneously detected by nanopores integrated with position-encoded DNA carriers.^{48,49,51,86,87} As shown in Fig. 7, the encoded DNA probe mainly involves a double-stranded DNA carrier containing protein-binding aptamers or oligonucleotides complemented with viral genes at specific positions.⁴⁹ Varied antigen proteins, including spike and nucleocapsids, are directly detected in raw human saliva, and multiple gene fragments from clinical samples are discriminated among viral variants of SARS-CoV-2 in a single test. The programmable DNA carriers are encoded with specific binding sites along their scaffold strand, which provides the possibility of simultaneous detection of multiple targets involving proteins and nucleic acids, and so on. The design greatly improves the sensitivity and specificity of nanopore sensors for multiplexed detection. Specifically, DNA carriers with uniform negative charges tend to enter nanopores under the control of electric fields, and produce deeper current blockade signals with the increasing molecular volume and spatial flexibility of DNA nanostructures. The target signals can be amplified and discriminated from interference within complex mixtures. Thus, the nanopores assisted by encoded DNA probes greatly improve the precision diagnosis of multiple viruses in a single assay, which is more suitable for point-of-care testing.

Fig. 7 Schematic showing ion current–time traces for the multiplexed sensing of surface proteins and RNA fragments of SARS-CoV-2, and the long dsDNA probes encoded with S/N protein-binding aptamer and viral RNA fragments.⁴⁹ Reproduced from ref. 49 with permission from Springer Nature Limited, copyright 2023.

6. Virus detection by nanopores integrated with other technologies

As described above, the capability of nanopores can be enhanced with the integration of a variety of technologies, such as molecular probes, nanostructures and signal algorithms for sample preparation, detection, and data analysis.^53–58 Typically, Liu et al. developed Au nanoparticles labeled CRISPR-Cas13a assay for target RNA detection of SARS-CoV-2 viruses by nanopore platforms.⁵⁷ As shown in Fig. 8(A), the viral RNA is recognized by the CRISPR-Cas13a assay since the Cas13a:crRNA complex can target the specific nucleotide sequence of the S reading frame of SARS-CoV-2. Then, streptavidin-coated Au nanoparticles are added to the assay and allowed to interact with biotin-FAM-labeled ssRNA probes. The addition of target RNA activates the Cas13a:crRNA complex to cleave the ssRNA probes in the mixtures. The uncleaved ssRNA probes are isolated by magnetic beads, and the CRISPR-Cas13-cleaved Au nanoparticles are introduced into nanopore sensors for quantitative detection. The amount of measured Au nanoparticles can be used for the quantitative sensing of RNA targets. The novel strategy of nanopore sensing integrated with multiple technologies can discriminate SARS-CoV-2 from SARS-1 at levels as low as the femtomolar level, ideally, for the accurate detection of low virus samples. Similarly, some two-dimensional nanomaterials, like graphene, are also employed to functionalize solid-state nanopores for the sensitive detection of the envelope glycoprotein of HIV.⁸⁸ The graphene-based nanomaterials improve the sensitivity of nanopores by basically changing the ionic transport properties across the mesoporous membranes.

Fig. 8 (A) Schematic of the AuNP-Cas13-based nucleic acid detection strategy by nanopore platforms, including Cas13a/crRNA complex cleavage of ssRNA probes, streptavidin-coated AuNPs binding biotin-labeled ssRNA probes, and the product capture and isolation by magnetic beads.⁵⁷ (B) Schematic of the nanopore-gated optofluidic device, and the representative electrical and optical signals corresponding to single-particle detection events.⁵⁴ Reproduced from ref.57 with permission from WILEY-VCH Verlag GmbH & Co., copyright 2022. Reproduced from ref.54 with permission from the American Chemical Society, copyright 2014.

Microfluidics and optofluidics have also garnered a lot of attention during nanopore development. Harms et al. designed a nanofluidic channel containing two holes arranged successively to measure the electrophoretic mobility and particle size of single hepatitis B virus (HBV) capsids.⁸⁹ Tsutsui et al. studied high-throughput virus detection utilizing microfluidics control and multi-pore electrophoresis in a nanopore system integrated on a thin silicon nitride membrane.⁵³ Sampad et al. developed a combined optofluidic-nanopore platform for RNA quantitative analysis of Zika and SARS-CoV-2 under label-free and amplification-free circumstances. The assay is achieved at an atto-molar limit of detection in different clinical biofluid samples from semen, urine, and swabs. Typically, Liu et al. reported a dual-mode sensing platform acquiring electrical and optical signals synchronously by nanopores integrated with an optofluidic chip.⁵⁴ As shown in Fig. 8(B), each virus particle enters the nanopore, generating a current blockade. Meanwhile, nanopores serve as intelligent gates to control the motion of individual viral particles into the optical excitation region, thereby enabling rapid and continuous label-free optical analysis. Both optical and electrical signals on nanopore-gated optofluidic devices are highly correlated for virus particle identification in rapid succession with high fidelity. The synchronized electro-optical sensing is a promising method for providing accurate, complete and reliable information about more analytes towards clinical samples.

7. Summary and outlook

In summary, the detection and quantitative analysis of massive viruses and molecular targets have been explored based on the rapid development and gradual integration of nanopore sensors with other biotechnologies, to resist the growing threat of viral outbreaks to human health. These novel intelligent methods have been summarized in Table 1 based on the nanopore characteristics and virus targets.

Table 1 Nanopore-based virus detection

Nanopores	Targets	Advantages	Ref.
Oxford nanopore	SARS-CoV-2 genome	SNVs detected at >99% sensitivity and >99% precision above a minimum ∼60-fold coverage depth.	18
Oxford nanopore	SARS-CoV-2, influenza A and B, parainfluenza, respiratory syncytial virus, and rhinovirus	Simultaneously detecting SARS-CoV-2 and other respiratory viruses within 6–10 h, with a limit of detection of ten standard plasmid copies per reaction.	12
Silicon nitride nanopores	Filamentous virus fd	A voltage-independent mobility, obeying first-passage-time statistics.	32
Peptide-decorated Au/Si3N4 nanopore	Influenza A (H1N1), B	Single-virus discriminations via surface–peptide Interactions.	37
DNA aptamer-PET nanopore	Adenovirus/SARS-CoV-2	Sensitivity down to 1 pfu ml^−1 for human adenovirus and 1 × 10^4 copies per ml for SARS-CoV-2.	36
Nanopore with machine learning	Influenza A and B, RSV, adenovirus, coronavirus	Over 99% accuracy for influenza A and B, RSV, adenovirus, coronavirus.	34
Nanopore with machine learning	HCoV-229E, SARS-CoV, MERS-CoV, SARS-CoV-2	Identifying coronaviruses similar in size, detection of SARS-CoV-2 in saliva specimens with a sensitivity of 90% and specificity of 96% with 5-minute measurement.	33
Silicon nitride nanopores	HIV and EBV viruses	Direct and precise quantitative measurement of particles’ zeta-potentials.	72
Silicon nitride nanopores	Adeno-associated virus (AAV)	Sensing AAV vectors packaged with DNA of different lengths.	69
Au-SiN nanopores with optical detection	Human immunodeficiency virus, murine leukemia virus, adeno-associated virus, hepatitis B virus	A jamming phenomenon associated with virus confinement under flow.	75
α-Hemolysin	Influenza A viruses	Detection of conserved influenza A virus RNA promoter.	52
Silicon nitride nanopores	Human immunodeficiency virus (HIV-1B).	Detection of conserved viral sequences and other nucleic acid motifs.	45
RT-LAMP coupled glass nanopores	Coronavirus SARS-CoV-2	LOD of 65 copies at the 95% confidence level.	44
Cytolysin A nanopore	Influenza Virus	Probing neuraminidase activity of the influenza virus, LOD of 0.17 ng mL^−1.	78
Silicon nitride nanopores	HIV-1 virus	Detection of the nucleocapsid protein (NCp7) of the human immunodeficiency virus 1 (HIV-1).	47
Glass nanopore	SARSCoV-2	Multiplexed detection of viral antigen and RNA of SARSCoV-2 lineages of wild-type B.1.1.7 (Alpha), B.1.617.2 (delta), and B.1.1.539 (omicron).	49
AuNP-Cas13-based nanopores	SARS-CoV-2	A detection limit of 50 × 10^−15 m (30 000 copies μl^−1).	57
Nanopore-gated optofluidic chip	H1N1 influenza A viruses	Electrical and optical analysis up to 100% fidelity.	54

---

Viral sequencing by nanopores provides high-resolution molecular insights into deciphering viral pathogenesis, mutational evolution and transmission modes, owing to its unique advantages of long-length reading and free labeling. The commercial devices of ONT sequencing have been widely used for Ebola, Zika, SARS-CoV-2 and other disease outbreaks. However, nanopore sequencing remains at an early stage of commercial development, and the sequencers still encounter the issues of flow cells with blocked or functionality-loss pores. Although the average sequencing accuracy is improving, certain fragments of sequence reading have an inferior base calling accuracy, especially DNA/RNA modifications, short indels and variants at low read-count frequencies. To acquire high-quality sequencing raw data, the targeted gene enrichment is still required through the optimization of gene extraction and genomic library construction for low-abundance pathogens from host samples. Meanwhile, more opportunities remain for improving bioinformatic analysis involving base calling algorithms, genome assembly, and adapting software to long reads to reach single-base accuracy comparable to other sequencing platforms.

With the increasing demand for pathogen surveillance and clinical applications, single-entity analyses of viruses and the related biomarkers are expected to provide real-time and accurate virus diagnosis and drug screening. Size-adjustable solid-state nanopores are the preferred choice for virus counting, providing a rapid, low-cost and portable platform for the label-free electrical diagnostics of viral pathogens. The confined space of nano-scale pores enables them to sense more closely the physical features of single viruses for classifying virus species and quantifying virus particles with shape, surface charge, and biological activity. Although existing experiments indicate that the molecular capture is related to the concentration of viruses, it is hard to avoid the deformation, membrane fusion, and jamming of larger viral particles in nanopore transport, and non-specific interactions during virus translocation also interfere with target identification, with the possibility of false positives and false negatives. In response to this, some protein or nucleic acid molecular probes have been designed and anchored within nanopores to enhance the bio-selectivity on viruses, and the emerging machine learning algorithms have also made it possible to discriminate subtle changes of resistive pulses from particle counting. Large amounts of high-quality signals are necessary for data training and particle quantitative analysis. A standardized database needs to be constructed through extensive signal collection and feedback correction from various virus samples.

Sensitive detection mainly depends on the susceptibility of ionic channel. The differences in nanopore structure and size can affect the reliability and repeatability of current signals. Pore enlargement and fouling may occur in nanopore experiments because various molecules and impurities are deposited and eroded in the nanoscale channels. To achieve reliable long-term viral detection, more durable solid nanopores with higher precision and stability need to be designed, maintaining nanopore performance reproducibility. Recent studies demonstrate that polymer coatings and zwitterionic monolayers can significantly mitigate pore clogging issues.^90,91 Thus, functional nanopore sensors assisted by various molecular probes and artificial intelligence reduce the uncertainty of nanopore geometry and particle interactions, and have the potential to be portable platforms for the detection of virus infection in on-site rapid diagnostics.

Besides particle counting of whole viruses, the nanopore-based single-molecular detection is also an alternative non-sequencing approach for rapid, cost-effective and accurate viral diagnostics. So far, various molecular biomarkers, including some related nucleic acids/proteins of viral samples, have been developed and utilized in pandemic surveillance, such as SARS-CoV-2, HIV and so on. Individual molecules with more refined structures and interacting sites are suitable for ultra-sensitivity and real-time detection by both biological and solid nanopores, which also provide more important physiological information about virus replication, invasion patterns and drug screening. The sensitivity of single-molecular detection is well documented from the previous nanopore experiments, and one of the key issues is the target design. The ideal molecules should be selected with high specificity and isolated from the clinical samples. Particularly, low-abundance target samples in clinical settings need to be enriched and captured by nanopores combined with molecular biotechnologies, nanomaterials and microfluidic systems.

To enhance the diagnostic accuracy, multiple biomarkers are preferred over single species. Moreover, the direct detection of biomarkers from body fluids (blood, tears, urine, etc.) remains a challenge for nanopore measurements. To address these issues, one recent trend is to integrate nanopore sensors with other technologies for nanopore optimization, sample preparation, and data analysis. As mentioned above, molecular probes, nanomaterials, and microfluidics have all improved the precision and specificity of the nanopore platform, although they also increase the complexity of devices and operations.

In contrast, machine learning and more signal processing algorithms seem to be more adaptive for nanopore sensing. Firstly, statistical analysis of massive nanopore signals is a common requirement for sequencing and single-molecule/particle sensing, reducing dependence on sample preparation and dedicated instruments. Secondly, deep learning algorithms have the potential to evaluate and optimize experiment design to improve signal quality. An automated analysis platform is also necessary with the aid of artificial intelligence, without dependency on the trained staff. With continuous technological innovation and interdisciplinary collaboration, the co-design of experiments and algorithms will ultimately establish rapid and ultra-sensitive nanopore platforms for the digital diagnosis of virus infection, offering strong support for public health safety.

Conflicts of interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

No data, models, or code were generated or used during the study.

Acknowledgements

This work was supported by the National Key R&D Program of China (No. 2024YFF1206201) and the National Natural Science Foundation of China (62371128).

References

1. S. Moutinho, Science, 2024, 384, 1052–1053.
2. A. J. Eisfeld, A. Biswas, L. Guan, C. Gu, T. Maemura, S. Trifkovic, T. Wang, L. Babujee, R. Dahn and P. J. Halfmann, Nature, 2024, 633, 426–432.
3. I. C. Chow and S.-S. Wong, Nat. Immunol., 2024, 25, 1319–1321.
4. J. Lim, K. Koprowski, R. Stavins, N. Xuan, T.-H. Hoang, J. Baek, V. Kindratenko, L. Khaertdinova, A. Y. Kim and M. Do, ACS Sens., 2024, 9, 4058–4068.
5. N. Kumar, N. P. Shetti, S. Jagannath and T. M. Aminabhavi, Chem. Eng. J., 2022, 430, 132966.
6. M. M. Lamers and B. L. Haagmans, Nat. Rev. Microbiol., 2022, 20, 270–284.
7. H.-J. Rheinberger, Science, 2000, 289, 731.
8. Y. Qiu, Q. Li, B. E. Lee, N. J. Ruecker, N. F. Neumann, N. J. Ashbolt and X. Pang, Water Res., 2018, 147, 73–81.
9. P. K. Olitsky and L. Boëz, J. Exp. Med., 1927, 45, 685–699.
10. V. Servellita, A. Sotomayor Gonzalez, D. M. Lamson, A. Foresythe, H. J. Huh, A. L. Bazinet, N. H. Bergman, R. L. Bull, K. Y. Garcia and J. S. Goodrich, Nature, 2023, 617, 574–580.
11. E. Norrby, G. Biberfeld, F. Chiodi, A. von Gegerfeldt, A. Nauclér, E. Parks and R. Lerner, Nature, 1987, 329, 248–250.
12. M. Wang, A. Fu, B. Hu, Y. Tong, R. Liu, Z. Liu, J. Gu, B. Xiang, J. Liu and W. Jiang, Small, 2020, 16, 2002169.
13. D. Tombácz, Á. Dörmő, G. Gulyás, Z. Csabai, I. Prazsák, B. Kakuk, Á. Harangozó, I. Jankovics, B. Dénes and Z. Boldogkői, GigaScience, 2022, 11, 1–11.
14. X. Deng, A. Achari, S. Federman, G. Yu, S. Somasekar, I. Bártolo, S. Yagi, P. Mbala-Kingebeni, J. Kapetshi and S. Ahuka-Mundeke, Nat. Microbiol., 2020, 5, 443–454.
15. T. Charalampous, G. L. Kay, H. Richardson, A. Aydin, R. Baldan, C. Jeanes, D. Rae, S. Grundy, D. J. Turner and J. Wain, Nat. Biotechnol., 2019, 37, 783–792.
16. A. L. Greninger, S. N. Naccache, S. Federman, G. Yu, P. Mbala, V. Bres, D. Stryke, J. Bouquet, S. Somasekar and J. M. Linnen, Genome Med., 2015, 7, 1–13.
17. S. L. Yang, L. DeFalco, D. E. Anderson, Y. Zhang, J. G. A. Aw, S. Y. Lim, X. N. Lim, K. Y. Tan, T. Zhang and T. Chawla, Nat. Commun., 2021, 12, 5113.
18. R. A. Bull, T. N. Adikari, J. M. Ferguson, J. M. Hammond, I. Stevanovski, A. G. Beukers, Z. Naing, M. Yeang, A. Verich and H. Gamaarachchi, Nat. Commun., 2020, 11, 6272.
19. Y. Che, Z. Li, S. Xie and C. Chen, Curr. Nanosci., 2024, 20, 157–173.
20. S. Akhtarian, S. Miri, A. Doostmohammadi, S. K. Brar and P. Rezai, Bioengineered, 2021, 12, 9189–9215.
21. C.-M. Ji, X.-Y. Feng, Y.-W. Huang and R.-A. Chen, Viruses, 2024, 16, 798.
22. J. P. Fried, J. L. Swett, B. P. Nadappuram, J. A. Mol, J. B. Edel, A. P. Ivanov and J. R. Yates, Chem. Soc. Rev., 2021, 50, 4974–4992.
23. X. Chen, S. Zhou, Y. Wang, L. Zheng, S. Guan, D. Wang, L. Wang and X. Guan, TrAC, Trends Anal. Chem., 2023, 162, 117060.
24. S. Buddle, L. Forrest, N. Akinsuyi, L. M. Martin Bernal, T. Brooks, C. Venturini, C. Miller, J. R. Brown, N. Storey and L. Atkinson, Genome Med., 2024, 16, 111.
25. Y. Liu, J. Kearney, M. Mahmoud, B. Kille, F. J. Sedlazeck and T. J. Treangen, Nat. Commun., 2022, 13, 1321.
26. D. Kim, J.-Y. Lee, J.-S. Yang, J. W. Kim, V. N. Kim and H. Chang, Cell, 2020, 181, 914–921.e910.
27. D. J. Baker, A. Aydin, T. Le-Viet, G. L. Kay, S. Rudder, L. de Oliveira Martins, A. P. Tedim, A. Kolyva, M. Diaz and N.-F. Alikhan, Genome Med., 2021, 13, 1–11.
28. J. Quick, N. J. Loman, S. Duraffour, J. T. Simpson, E. Severi, L. Cowley, J. A. Bore, R. Koundouno, G. Dudas and A. Mikhail, Nature, 2016, 530, 228–232.
29. S. Van Puyvelde and S. Argimon, Nat. Rev. Microbiol., 2019, 17, 5.
30. B. Kakuk, Á. Dörmő, Z. Csabai, G. Kemenesi, J. Holoubek, D. Růžek, I. Prazsák, V. É. Dani, B. Dénes and G. Torma, Sci. Data, 2023, 10, 262.
31. A. G. Shaw, T. K. Mampuela, E. L. Lofiko, C. Pratt, C. Troman, E. Bujaki, Á. O’Toole, J. O. Akello, A. A. Aziza and E. K. Lusamaki, Nat. Microbiol., 2023, 8, 1634–1640.
32. A. McMullen, H. W. De Haan, J. X. Tang and D. Stein, Nat. Commun., 2014, 5, 4171.
33. M. Taniguchi, S. Minami, C. Ono, R. Hamajima, A. Morimura, S. Hamaguchi, Y. Akeda, Y. Kanai, T. Kobayashi and W. Kamitani, Nat. Commun., 2021, 12, 3726.
34. A. Arima, M. Tsutsui, T. Yoshida, K. Tatematsu, T. Yamazaki, K. Yokota, S. I. Kuroda, T. Washio, Y. Baba and T. Kawai, ACS Sens., 2020, 5, 3398–3403.
35. K. Murakami, S. I. Kubota, K. Tanaka, H. Tanaka, K. Akabane, R. Suzuki, Y. Shinohara, H. Takei, S. Hashimoto and Y. Tanaka, Lab Chip, 2023, 23, 4909–4918.
36. A. S. Peinetti, R. J. Lake, W. Cong, L. Cooper, Y. Wu, Y. Ma, G. T. Pawel, M. E. Toimil-Molares, C. Trautmann and L. Rong, Sci. Adv., 2021, 7, eabh2848.
37. A. Arima, I. H. Harlisa, T. Yoshida, M. Tsutsui, M. Tanaka, K. Yokota, W. Tonomura, J. Yasuda, M. Taniguchi and T. Washio, J. Am. Chem. Soc., 2018, 140, 16834–16841.
38. A. Arima, M. Tsutsui, I. H. Harlisa, T. Yoshida, M. Tanaka, K. Yokota, W. Tonomura, M. Taniguchi, M. Okochi and T. Washio, Sci. Rep., 2018, 8, 16305.
39. L. Zhao, Y. Deng, Y. Wang, S. Zhou, B. Yin, Y. Chen, Y. Wang, J. Li, L. Wang and Y. Lin, Mater. Today Phys., 2024, 46, 101479.
40. M. Dong, A. Kshirsagar, A. J. Politza and W. Guan, ACS Nano, 2024, 18, 7170–7179.
41. R. Nouri, Y. Jiang, X. L. Lian and W. Guan, ACS Sens., 2020, 5, 1273–1280.
42. R. Nouri, Y. Jiang, Z. Tang, X. L. Lian and W. Guan, Nano Lett., 2021, 21, 8393–8400.
43. M. A. Ahamed, M. A. U. Khalid, M. Dong, A. J. Politza, Z. Zhang, A. Kshirsagar, T. Liu and W. Guan, Biosens. Bioelectron., 2024, 246, 115866.
44. Z. Tang, R. Nouri, M. Dong, J. Yang, W. Greene, Y. Zhu, M. Yon, M. S. Nair, S. V. Kuchipudi and W. Guan, Biosens. Bioelectron., 2022, 197, 113759.
45. K. Sethi, G. P. Dailey, O. K. Zahid, E. W. Taylor, J. A. Ruzicka and A. R. Hall, ACS Nano, 2021, 15, 8474–8483.
46. Y. Horiguchi, N. Naono, O. Sakamoto, H. Takeuchi, S. Yamaoka and Y. Miyahara, ACS Appl. Mater. Interfaces, 2022, 14, 20168–20178.
47. D. J. Niedzwiecki, R. Iyer, P. N. Borer and L. Movileanu, ACS Nano, 2013, 7, 3341–3350.
48. S. Cai, R. Ren, J. He, X. Wang, Z. Zhang, Z. Luo, W. Tan, Y. Korchev, J. B. Edel and A. P. Ivanov, Nano Lett., 2023, 23, 11438–11446.
49. R. Ren, S. Cai, X. Fang, X. Wang, Z. Zhang, M. Damiani, C. Hudlerova, A. Rosa, J. Hope and N. J. Cook, Nat. Commun., 2023, 14, 7362.
50. Y. Zhang, X. Chen, C. Wang, H.-C. Chang and X. Guan, Biosens. Bioelectron., 2022, 196, 113697.
51. F. Bošković, J. Zhu, R. Tivony, A. Ohmann, K. Chen, M. F. Alawami, M. Đorđević, N. Ermann, J. Pereira-Dias and M. Fairhead, Nat. Nanotechnol., 2023, 18, 290–298.
52. S. Oh, M.-K. Lee and S.-W. Chi, ACS Sens., 2019, 4, 2849–2853.
53. M. Tsutsui, T. Yamazaki, K. Tatematsu, K. Yokota, Y. Esaki, Y. Kubo, H. Deguchi, A. Arima, S. I. Kuroda and T. Kawai, Nanoscale, 2019, 11, 20475–20484.
54. S. Liu, Y. Zhao, J. W. Parks, D. W. Deamer, A. R. Hawkins and H. Schmidt, Nano Lett., 2014, 14, 4816–4820.
55. M. J. N. Sampad, H. Zhang, T. D. Yuzvinsky, M. A. Stott, A. R. Hawkins and H. Schmidt, Biosens. Bioelectron., 2021, 194, 113588.
56. M. J. N. Sampad, S. Saiduzzaman, Z. J. Walker, T. N. Wells, J. X. Wayment, E. M. Ong, S. D. Mdaki, M. A. Tamhankar, T. D. Yuzvinsky and J. L. Patterson, Proc. Natl. Acad. Sci. U. S. A., 2024, 121.
57. L. Liu, Z. Xu, K. Awayda, S. J. Dollery, M. Bao, J. Fan, D. Cormier, M. R. O'Connell, G. J. Tobin and K. Du, Adv. Mater. Technol., 2022, 7, 2101550.
58. J. Reitemeier, J. Metro and K. X. Fu, Sens. Actuators Rep., 2024, 8, 100225.
59. A. J. McMullen, J. X. Tang and D. Stein, ACS Nano, 2017, 11, 11669–11677.
60. M. Chinappi, M. Yamaji, R. Kawano and F. Cecconi, ACS Nano, 2020, 14, 15816–15828.
61. M. Tsutsui, Y. Tsunekawa, M. Wada, A. Arima, A. Onodera, M. Nishina, M. Nagoya, Y. Baba, T. Kawai and T. Okada, Small Methods, 2025, 2401321.
62. W. Tang, J. P. Fried, R. D. Tilley and J. J. Gooding, Chem. Soc. Rev., 2022, 51, 5757–5776.
63. Z. L. Hu, M. Z. Huo, Y. L. Ying and Y. T. Long, Angew. Chem., 2021, 133, 14862–14873.
64. Y. Wang, Y. Zhao, A. Bollas, Y. Wang and K. F. Au, Nat. Biotechnol., 2021, 39, 1348–1365.
65. Y.-L. Ying, Z.-L. Hu, S. Zhang, Y. Qing, A. Fragasso, G. Maglia, A. Meller, H. Bayley, C. Dekker and Y.-T. Long, Nat. Nanotechnol., 2022, 17, 1136–1146.
66. Y. Wang, Y. Zhao, A. Bollas, Y. Wang and K. F. Au, Nat. Biotechnol., 2021, 39, 1348–1365.
67. J. D. Uram, K. Ke, A. J. Hunt and M. Mayer, Small, 2006, 2, 967–972.
68. A. Arima, M. Tsutsui, T. Washio, Y. Baba and T. Kawai, Anal. Chem., 2020, 93, 215–227.
69. M. Tsutsui, M. Wada, A. Arima, Y. Tsunekawa, T. Sasaki, K. Sakamoto, K. Yokota, Y. Baba, T. Kawai and T. Okada, ACS Nano, 2024, 18, 15695–15704.
70. K. Zhou, L. Li, Z. Tan, A. Zlotnick and S. C. Jacobson, J. Am. Chem. Soc., 2011, 133, 1618–1621.
71. B. I. Karawdeniya, Y. N. D. Bandara, A. I. Khan, W. T. Chen, H.-A. Vu, A. Morshed, J. Suh, P. Dutta and M. J. Kim, Nanoscale, 2020, 12, 23721–23731.
72. N. Arjmandi, W. Van Roy, L. Lagae and G. Borghs, Anal. Chem., 2012, 84, 8490–8496.
73. N. Arjmandi, W. Van Roy and L. Lagae, Anal. Chem., 2014, 86, 4637–4641.
74. I. Phase, I. I. I. P. IIPhase and I. I. I. I. I. Phase, Nat. Rev. Drug Discovery, 2021, 20, 173.
75. L. Chazot-Franguiadakis, J. Eid, G. Delecourt, P. J. Kolbeck, S. Brugère, B. Molcrette, M. Socol, M. Mougel, A. Salvetti and V. Démery, Nat. Commun., 2024, 15, 6180.
76. I. M. Mackay, K. E. Arden and A. Nitsche, Nucleic Acids Res., 2002, 30, 1292–1305.
77. N. S. Galenkamp, M. Soskine, J. Hermans, C. Wloka and G. Maglia, Nat. Commun., 2018, 9, 4085.
78. D.-K. Kwak, J.-S. Kim, M.-K. Lee, K.-S. Ryu and S.-W. Chi, Anal. Chem., 2020, 92, 14303–14308.
79. X. Zhou, B. Chu, R. Tang, Y. Jiang, D. Xi and S. Zhang, Sens. Actuators, B, 2024, 409, 135634.
80. J. G. Levin, J. Guo, I. Rouzina and K. Musier-Forsyth, Prog. Nucleic Acid Res. Mol. Biol., 2005, 80, 217–286.
81. Q. Liu, Y. Wang, Y. Liu, H. Wang, W. Li, P. Tang, T. Weng, S. Zhou, L. Liang and J. Yuan, Nanoscale, 2020, 12, 19711–19718.
82. L. Ratinho, N. Meyer, S. Greive, B. Cressiot and J. Pelta, Nat. Commun., 2025, 16, 3211.
83. S. Straathof, G. Di Muccio and G. Maglia, J. Am. Chem. Soc., 2025, 147, 15050–15065.
84. S. Straathof, G. Di Muccio, M. Yelleswarapu, M. Alzate Banguero, C. Wloka, N. J. van der Heide, M. Chinappi and G. Maglia, ACS Nano, 2023, 17, 13685–13699.
85. P. Fan, S. Zhang, Y. Wang, T. Li, H. Zhang, P. Zhang and S. Huang, Nat. Commun., 2024, 15, 1970.
86. Z. Zhang, X. Wang, X. Wei, S. W. Zheng, B. J. Lenhart, P. Xu, J. Li, J. Pan, H. Albrecht and C. Liu, Biosens. Bioelectron., 2021, 181, 113134.
87. U. F. Keyser, Nat. Nanotechnol., 2016, 11, 106–108.
88. A. Nehra, W. Chen, D. S. Dimitrov, A. Puri and K. P. Singh, ACS Appl. Mater. Interfaces, 2017, 9, 32621–32634.
89. Z. D. Harms, D. G. Haywood, A. R. Kneller, L. Selzer, A. Zlotnick and S. C. Jacobson, Anal. Chem., 2015, 87, 699–705.
90. J.-W. Seo, K. Fu, S. Correa, M. Eisenstein, E. A. Appel and H. T. Soh, Sci. Adv., 2022, 8, eabk2901.
91. Y. Chen, K. X. Fu, R. Cotton, Z. Ou, J. W. Kwak, J.-C. Chien, V. Kesler, H. Y. Y. Nyein, M. Eisenstein and H. Tom Soh, Nat. Biomed. Eng., 2025, 1–14.

---