Nanopore sequencing reads DNA or RNA by threading single molecules through a tiny protein or synthetic hole and measuring changes in electrical current as each base passes through. Unlike older sequencing methods that require chopping DNA into short fragments, copying it millions of times, and reading those copies optically, nanopore platforms analyze native molecules one at a time, in real time. The technology has upended assumptions about where and how quickly genomic analysis can happen, from hospital emergency wards to the International Space Station.
How the Technology Works
The basic setup involves a membrane with a nanometer-scale pore embedded in it. A voltage is applied across the membrane, creating an ionic current that flows through the pore. When a strand of DNA or RNA is drawn through by that voltage, it partially blocks the current. The amount of blockage varies depending on which nucleotide bases are inside the pore at any given moment, producing a distinctive electrical signal for each stretch of sequence. Because the molecule threads through in its natural order, the sequence of signals directly reflects the sequence of bases along the strand.
This single-molecule approach means there is no need for the amplification or labeling steps that most sequencing technologies require. Kilobase-length stretches of genomic DNA or RNA can be identified and characterized directly, which eliminates biases introduced by copying the DNA beforehand.1PubMed Central. The potential and challenges of nanopore sequencing The highly confined space inside the pore enforces what researchers call “perfect processivity,” meaning each base passes through in strict order and gets read sequentially rather than in parallel fragments that need to be reassembled computationally.
Biological Pores Versus Solid-State Pores
There are two broad categories of nanopores, and each has trade-offs. Biological nanopores use protein channels harvested from bacteria. Because they are produced by biological machinery, they are extraordinarily reproducible at the atomic level. Every pore protein folds into the same shape, which means the sensing geometry is consistent from one pore to the next. The commercially available Oxford Nanopore Technologies platform uses this approach, relying on engineered versions of bacterial pore proteins to do the actual sensing.
Solid-state nanopores, by contrast, are drilled into thin films of synthetic materials like silicon nitride or graphene. They offer advantages in structural stability and the ability to tune the pore diameter for specific applications. However, batch-to-batch variation remains a challenge because fabrication techniques cannot yet match the atomic-level precision of protein folding, although graphene-based pores are approaching that standard.2PubMed Central. Solid-State and Biological Nanopore for Real-Time Sensing of Single Chemical and Sequencing of DNA Biological pores excel at the molecular recognition needed for sequencing because specific amino acid interactions guide how the DNA sits inside the channel, while solid-state pores are better suited to applications where you need extreme durability or control over pore geometry.3PubMed. Probing nanopores: molecular dynamics insights into the mechanisms of DNA and protein translocation through solid-state and biological nanopores
Ultra-Long Reads and What They Unlock
The standout capability of nanopore sequencing is read length. Traditional short-read platforms produce reads of a few hundred bases, which then have to be stitched together computationally. Nanopore sequencers routinely produce reads tens of thousands of bases long, and specialized protocols have pushed individual reads beyond 800,000 bases. In a landmark human genome project, researchers generated ultra-long reads with a median length exceeding 100,000 bases and individual reads up to 882,000 bases.4Nature Biotechnology. Nanopore sequencing and assembly of a human genome with ultra-long reads
These extreme read lengths matter because genomes are full of repetitive sequences that confuse short-read assemblers. When a read is long enough to span an entire repeat and anchor into unique sequence on both sides, the assembly software can place it unambiguously. Adding just five-fold coverage of ultra-long nanopore reads to an existing dataset more than doubled the contiguity of a human genome assembly and enabled researchers to piece together the entire four-megabase major histocompatibility complex, a notoriously difficult region, in one continuous stretch.5Nature Biotechnology. Nanopore sequencing and assembly of a human genome with ultra-long reads The same reads allowed measurement of telomere repeat lengths and closure of gaps in the human reference genome.
The Accuracy Problem and How It Is Improving
For all its advantages in read length and portability, nanopore sequencing has historically lagged behind short-read platforms in raw accuracy. Error rates have typically ranged between about 5% and 15%, depending on the chemistry version and the computational tools used to interpret the signal.6Genome Biology. From squiggle to basepair: computational approaches for improving nanopore sequencing read accuracy Those errors are not random. Runs of the same base, called homopolymers, and regions with short tandem repeats together account for roughly half of all sequencing errors. These stretches produce similar current levels for consecutive identical bases, making it hard for the software to count exactly how many of that base are present.
The composition of the DNA also matters. Regions with high GC content tend to accumulate more errors than low-GC regions, with error rates around 8% versus 6% respectively in one analysis. The dominant error type in problematic regions is deletions, though some reads show long insertions as well.7PubMed Central. Sequencing DNA with nanopores: Troubles and biases
Researchers have attacked these errors from two directions. One is improving the chemistry and pore proteins themselves, which Oxford Nanopore has done through successive hardware iterations. The other is developing better computational methods for translating the raw electrical signal into base calls. Neural network basecallers have made large gains, but some systematic errors resist correction because they stem from the physics of how the DNA interacts with the pore rather than from random noise. A tool called Homopolish, for example, trains on homologous sequences rather than nanopore reads themselves, specifically targeting the systematic errors that standard polishing tools leave behind.8PubMed Central. Homopolish: a method for the removal of systematic errors in nanopore sequencing by homologous polishing
How It Stacks Up Against Illumina and PacBio
The three main sequencing platforms in use today are Illumina (short reads, very high accuracy), PacBio (long reads, high accuracy after circular consensus correction), and Oxford Nanopore (very long reads, lower per-read accuracy but rapidly improving). How they compare depends heavily on what you are trying to do.
For microbiome profiling, a comparison of all three platforms sequencing rabbit gut communities found that they performed similarly down to the family level, classifying at least 99% of sequences. Differences emerged at finer taxonomic scales. Oxford Nanopore classified 91% of sequences to genus level and 76% to species level, outperforming PacBio at 85% and 63% and Illumina at 80% and 47%.9Frontiers in Microbiomes. Comparative analysis of Illumina, PacBio, and nanopore for 16S rRNA gene sequencing of rabbit’s gut microbiota The long reads that span entire marker genes give nanopore a resolution advantage for telling closely related species apart.
For viral genomics, the picture flips. When researchers benchmarked all three technologies for recovering phage genomes, Illumina-only assemblies came out ahead in both genome completeness and error rates. The best nanopore assemblies had single-nucleotide errors roughly 41% higher and insertion/deletion errors 157% higher than Illumina assemblies. PacBio fell in between. The most effective strategy turned out to be combining Illumina and nanopore reads in a hybrid assembly, which brought error rates back down to short-read-only levels while preserving some of the structural advantages of long reads.10PubMed Central. The long and short of it: benchmarking viromics using Illumina, Nanopore and PacBio sequencing technologies
For bacterial pathogens, hybrid assembly tools that merge short and long reads have become the gold standard. Benchmarking studies of assembly software found that tools like Unicycler, which combine Illumina and nanopore data, consistently produced the most contiguous and complete genomes.11PubMed Central. Benchmarking hybrid assembly approaches for genomic analyses of bacterial pathogens using Illumina and Oxford Nanopore sequencing The practical takeaway is that nanopore rarely replaces Illumina outright. Instead, the two are increasingly used together, with long reads providing the structural scaffold and short reads polishing the fine details.
Real-Time Sequencing and Adaptive Sampling
One of the most distinctive features of nanopore sequencing is that data streams out in real time. You do not have to wait for a multi-day run to finish before seeing results. The instrument begins producing readable sequence data within minutes of loading a sample. This opens up a capability called adaptive sampling, where the sequencer decides on the fly whether to keep reading a particular strand or eject it from the pore. If the software recognizes that a strand belongs to a region of the genome you don’t care about, it reverses the voltage and spits it out, freeing the pore for a strand you do want.12PubMed Central. A comprehensive benchmarking of adaptive sampling tools for nanopore sequencing
Adaptive sampling is essentially a software-controlled enrichment of target sequences without any extra lab work. You can specify which chromosomal regions, genes, or organisms you want to focus on and let the device selectively sequence them while rejecting everything else. This is useful in clinical settings where you want to deeply sequence a particular gene panel but only have limited sample material, or in environmental monitoring where most of the DNA in a sample comes from organisms you are not interested in.
Outbreak Response and Field Deployment
The MinION, Oxford Nanopore’s smallest device, is roughly the size of a stapler and plugs into a laptop via USB. That portability has made nanopore sequencing the go-to tool for genomic surveillance in resource-limited settings. The most famous demonstration came during the 2015 Ebola outbreak in Guinea, where researchers packed the entire sequencing system into standard airline luggage, set it up in the field, and generated results less than 24 hours after receiving an Ebola-positive sample. The sequencing itself took as little as 15 to 60 minutes.13PubMed Central. Real-time, portable genome sequencing for Ebola surveillance
That field-ready capability proved its worth again during the COVID-19 pandemic, where portable long-read sequencing devices helped track viral evolution in settings where shipping samples to a central laboratory would have introduced unacceptable delays.14PubMed Central. Long-Read Sequencing for the Rapid Response to Infectious Diseases Outbreaks The ability to generate near-complete pathogen genomes on site, in hours rather than weeks, fundamentally changed how public health agencies could respond to emerging threats.
Diagnosing Infections at the Bedside
Beyond outbreak surveillance, nanopore sequencing is being tested as a clinical diagnostic tool, particularly for severe infections where conventional blood cultures are too slow or too insensitive. Blood cultures, the standard method, require the pathogen to grow in a bottle for hours to days, and many infections produce negative cultures despite clear clinical signs of sepsis.
In a prospective study of 40 patients admitted to an emergency ward with suspected bloodstream infections, nanopore metagenomic sequencing of cell-free DNA in blood plasma confirmed all pathogens found by standard blood culture and identified clinically relevant pathogens in an additional 11 patients whose cultures had come back negative.15PubMed Central. Application of rapid Nanopore metagenomic cell-free DNA sequencing to diagnose bloodstream infections: a prospective observational study A larger study of sepsis patients found that targeted nanopore sequencing achieved a 94% positivity rate compared to 30% for culture and was far better at detecting fungi, viruses, and atypical pathogens that culture methods often miss entirely.16PubMed Central. Clinical application of targeted nanopore sequencing in pathogen detection in patients with sepsis These are still proof-of-concept studies and have not replaced standard diagnostics, but they suggest a future in which sequencing complements culture for the sickest patients.
Reading RNA and Epigenetic Marks Directly
Most sequencing technologies require converting RNA into DNA before it can be read, a step called reverse transcription that introduces its own biases and loses information about chemical modifications on the original RNA molecule. Nanopore sequencers can thread native RNA directly through the pore, skipping that conversion entirely.17PubMed Central. Highly parallel direct RNA sequencing on an array of nanopores The current signal produced by a modified base differs slightly from the unmodified version, which means the same sequencing run that reads the RNA sequence can also flag chemical modifications like methylation.
The same principle applies to DNA methylation, a key epigenetic mark involved in gene regulation and disease. Because nanopore sequencing reads the native DNA strand, methylated cytosines produce a different current signature than unmethylated ones. A growing toolkit of methylation-calling software has been developed to extract this information from nanopore data, and systematic evaluations have compared seven such tools for human epigenome-wide studies.18Genome Biology. DNA methylation-calling tools for Oxford Nanopore sequencing: a survey and human epigenome-wide evaluation Getting base sequence and epigenetic status from a single experiment is something no other commercial sequencing platform can do natively.
Environmental DNA and Biodiversity Monitoring
Ecologists have embraced nanopore sequencing for environmental DNA studies, where DNA shed by organisms into water or soil is collected and sequenced to survey biodiversity without ever having to see or capture an animal. The portability of the MinION makes it possible to run the entire workflow in the field, from water sampling through DNA extraction, amplification, and sequencing, without a conventional laboratory.
A study in sub-Saharan Africa validated a full mobile laboratory workflow using nanopore sequencing at a lagoon site. Over six sampling days, the nanopore eDNA approach detected ten terrestrial vertebrate taxa, compared to seven detected by camera traps, with the two methods overlapping on four species. Nanopore sequencing also detected more local species overall than Illumina sequencing run on the same samples.19PLOS One. Evaluation of nanopore sequencing for increasing accessibility of eDNA studies in biodiverse countries
In tropical amphibian communities, researchers combined isothermal amplification with nanopore sequencing to build reference sequence libraries directly from skin swab samples collected in the field, then used those references to improve species-level identification from eDNA water samples.20PubMed Central. A Field-Deployable eDNA Metabarcoding Workflow Including De Novo Reference Assembly for Characterising Understudied Biodiversity Hotspots Similar field-portable workflows have been validated for reef fish monitoring in remote marine ecosystems, where the cost and logistics of shipping samples to distant sequencing centers would otherwise make routine biodiversity surveys impractical.21PubMed Central. Optimizing a Novel eDNA-Based Framework for Reef Fish Biodiversity Monitoring Using an Autonomous Filtration System and in situ Nanopore Sequencing
Sequencing in Space
If fieldwork in the tropics tests the limits of portable sequencing, the International Space Station tests them further. In 2016, NASA astronaut Kate Rubins performed the first DNA sequencing in space using a MinION aboard the ISS. The results showed that the platform’s performance was not adversely affected by transport to the station, loading in microgravity, or operation in that environment. Flow cells remained stable after six months in orbit, a duration comparable to a one-way trip to Mars.22Scientific Reports. Nanopore DNA Sequencing and Genome Assembly on the International Space Station
Earlier parabolic-flight experiments had already confirmed that the sequencer could produce reads during microgravity conditions, with the longest read in one experiment generated during a zero-gravity phase.23PubMed Central. Nanopore sequencing in microgravity Follow-up work pushed the envelope further, confirming that nanopore sequencing works at Mars-equivalent gravity, lunar gravity, and even near the extremely low gravity that would be experienced near Jupiter’s moon Europa. Performance remained consistent across different gravity levels, during dynamic accelerations, and despite significant vibrations.24PubMed Central. Nanopore sequencing at Mars, Europa, and microgravity conditions For future crewed missions, the ability to identify microbes and monitor crew health with a pocket-sized device is not a theoretical convenience; it is a practical necessity when the nearest diagnostic lab is millions of kilometers away.
The Data Processing Bottleneck
Generating sequence data in real time is only useful if you can analyze it in real time, and that remains a genuine bottleneck. The raw output of a nanopore sequencer is not a neat string of A, C, G, and T characters. It is a noisy analog electrical signal that must be interpreted by neural network basecallers, then aligned to reference genomes or assembled from scratch. These steps are computationally demanding, and running them on a laptop in the field can be slow enough to negate the speed advantage of the sequencer itself.25Nature Computational Science. Real-time raw signal genomic analysis using fully integrated memristor hardware Researchers are exploring hardware accelerators, including novel memory-based computing architectures, that could process the raw signal directly without the conventional basecalling step, potentially enabling true on-site, real-time genomic analysis without cloud computing or a large server.
Beyond DNA and Into Proteins
The most forward-looking application of nanopore sensing is not DNA or RNA at all, but proteins. Proteins are the molecules that carry out most of the work in cells, and characterizing them at the single-molecule level has long been harder than sequencing nucleic acids. Unlike DNA, proteins do not carry a uniform charge, they fold into complex three-dimensional shapes, and their twenty amino acid building blocks produce more subtle current differences than DNA’s four bases.
Despite those challenges, recent work has shown that biological nanopores can detect and distinguish peptides bearing different chemical modifications, including phosphorylation, nitration, and oxidation at specific positions and in various combinations. Deep learning models trained on the resulting current signatures can identify these modified peptides at concentrations in the picomolar range.26ACS Nano. Deep Learning-Assisted Single-Molecule Detection of Protein Post-translational Modifications with a Biological Nanopore A high-throughput nanopore sensing platform has coupled streamlined peptide preparation with AI-driven analysis to achieve accurate peptide differentiation and protein identification, including blinded studies where multiple proteins were correctly identified from their enzymatic digests.27Nature Communications. Nanopore-based massively parallel sensing for peptide profiling and protein identification
The ultimate goal is full single-molecule protein sequencing with single-amino-acid resolution. Researchers predict that nanopores will likely be capable of this, which would open doors to measuring the heterogeneity of protein modifications, quantifying rare proteins, and characterizing splicing variants that current proteomic technologies struggle with.28PubMed Central. Toward single-molecule protein sequencing using nanopores If that vision materializes, the same basic sensing principle that reads DNA today could become the foundation of a new generation of proteomic tools.