Target Audience

  • Entrepreneurs and R&D professionals interested in building businesses around affordable analysis of DNA sequences, chemical modifications such as methylation, and RNA molecules.
  • Founders and technology leaders seeking to understand the differences between Oxford Nanopore, PacBio, Illumina, Roche, and other sequencing platforms to select the most suitable technologies for their businesses.
  • Innovators exploring opportunities in genomics, epigenetics, RNA analysis, infectious disease diagnostics, drug discovery, and proteomics.

Executive Summary

  • Oxford Nanopore Technologies (ONT) has dramatically improved sequencing accuracy over the past decade and now offers unique advantages in ultra-long DNA sequencing, DNA methylation detection, and direct RNA analysis.
  • PacBio excels at highly accurate long-read sequencing, while Illumina, MGI, Element Biosciences, and Ultima Genomics are particularly competitive in high-throughput, cost-efficient sequencing of large numbers of samples.
  • Roche introduced its Sequencing by Expansion (SBX) technology commercially in 2026, using engineered molecules to improve electrical signal detection through nanopores.
  • Each technology has distinct strengths and limitations. Combining ONT with PacBio or short-read sequencing platforms can offer significant advantages for certain applications.
  • For entrepreneurs, the key consideration is not simply sequencing accuracy or cost per gigabase, but the types of molecular information that can be obtained, turnaround time, sample requirements, and total cost per successfully analyzed sample.

1. What Is Driving Competition in DNA Sequencing?

Approximately a decade ago, Oxford Nanopore Technologies (ONT) introduced a remarkable approach to DNA sequencing.

Instead of relying on optical detection, the company developed a technology that passes DNA molecules through microscopic protein pores, known as nanopores, and identifies DNA sequences from changes in electrical signals.

ONT launched its MinION early-access program in 2014 and began commercial availability in May 2015. [1]

At the time, the technology attracted considerable attention because it enabled DNA sequencing with an extraordinarily small device, unlike conventional large-scale sequencing instruments.

More than a decade later, however, ONT has evolved far beyond being simply a manufacturer of portable DNA sequencers.

Its technology can now read exceptionally long DNA molecules and provide information about DNA methylation and other chemical modifications, RNA sequences, and potentially RNA modifications.

In other words, sequencing is becoming a way to obtain multiple layers of biological information from individual molecules.

Meanwhile, competitors such as PacBio, Illumina, MGI, Element Biosciences, and Ultima Genomics have continued to improve sequencing accuracy, throughput, and cost efficiency through fundamentally different approaches.

In 2026, Roche also introduced a new sequencing platform that uses nanopores but relies on a substantially different method of preparing molecules for electrical detection.

The DNA sequencing market is therefore no longer simply a competition over which company can achieve the highest accuracy.

Instead, technologies based on different physical principles are developing advantages in different applications.

Three competitive dimensions are particularly important:

  1. How accurately can a technology read long DNA molecules?
  2. How quickly and inexpensively can it analyze large numbers of DNA or RNA samples?
  3. Can it detect molecular information beyond DNA sequences, including chemical modifications and RNA characteristics?

Understanding these differences is essential for entrepreneurs seeking to build new businesses around genomic information.

2. How Has Oxford Nanopore Technologies Evolved Over the Past Decade?

2.1. The Fundamental Principle of Nanopore Sequencing

The basic principle of ONT sequencing has remained largely unchanged.

A voltage is applied across a protein nanopore, generating an ionic current.

As a DNA molecule passes through the pore, its presence alters that current.

By analyzing these current changes, the system estimates the underlying DNA sequence.

Strictly speaking, the electrical signal is not determined by a single nucleotide alone.

Several nucleotides located near the nanopore's sensing region contribute simultaneously to the observed current.

Machine-learning models interpret these complex electrical signals and translate them into sequences of A, T, G, and C.

This process is called basecalling.

The movement of DNA through the nanopore is also regulated by a motor protein, allowing the system to measure electrical signals as the molecule progresses through the pore.

The key distinction from conventional optical sequencing is that ONT directly measures electrical signals generated by individual nucleic acid molecules rather than detecting fluorescent labels.

Because chemical modifications such as DNA methylation can also influence these electrical signals, nanopore sequencing can potentially obtain sequence and modification information simultaneously. [2]

2.2. Technological Progress Over the Past Ten Years

ONT has continuously improved its nanopores, library preparation chemistry, sequencing hardware, and basecalling algorithms.

In particular, the combination of R10.4.1 nanopores, Kit V14 chemistry, and Dorado basecalling has substantially improved sequencing accuracy, which was previously one of the technology's major limitations. [2]

Technology Parameter Approximately 10 Years Ago As of 2026
Single-read accuracy Generally in the 80–90% range Approximately 99.75% (Q26) under representative conditions with advanced models
Read length Several kilobases to tens of kilobases Tens of kilobases routinely; over 100 kb and even megabase-scale reads under suitable conditions
Instrument platforms Primarily MinION From MinION to large-scale PromethION systems
Data output Relatively limited Supports large-scale applications, including human whole-genome sequencing
Modified-base detection Limited Detection of modifications such as 5mC and 5hmC from electrical signals
RNA sequencing Primarily cDNA-based analysis Direct RNA sequencing of native RNA molecules
Computational processing Early-generation basecalling High-accuracy Dorado models and GPU-based analysis
Real-time control Primarily data acquisition Adaptive Sampling, including rejection of unwanted molecules during sequencing

ONT's advanced Dorado basecalling models have demonstrated approximately 99.75% single-read accuracy (Q26) under representative conditions. [2]

One important distinction is the difference between single-read accuracy and consensus accuracy.

Single-read accuracy refers to the accuracy obtained when reading an individual DNA molecule.

Consensus accuracy refers to the accuracy obtained by combining information from multiple reads and correcting errors.

With sufficient sequencing coverage and appropriate analytical methods, ONT has demonstrated consensus accuracy exceeding Q50 in certain applications. [2]

Nevertheless, challenges remain in regions containing repetitive sequences, particular nucleotide compositions, and applications requiring detection of very low-frequency variants.

2.3. The Unique Value of Long Molecules and Chemical Modifications

One of ONT's most important capabilities is reading extremely long DNA molecules.

Reads exceeding 100 kb are possible, and reads longer than 1 Mb can be obtained under suitable experimental conditions.

This capability makes it possible to investigate complex genomic structures that cannot be resolved using short DNA fragments.

Another major advantage is the detection of DNA methylation without requiring the chemical conversion procedures commonly used in conventional methylation sequencing.

ONT also supports Direct RNA Sequencing, in which native RNA molecules are passed through nanopores without first converting them into complementary DNA.

Consequently, ONT has evolved from a DNA sequencing device into a platform capable of extracting multiple types of molecular information from nucleic acids.

3. Comparing the Major DNA Sequencing Technologies

The following table summarizes the major companies and their technologies.

Company / Technology Typical Read Length Key Strengths Main Limitations
Oxford Nanopore / Nanopore sequencing 10–100 kb, with ultra-long reads possible Ultra-long DNA, structural variants, direct RNA sequencing, rapid analysis Per-base errors and dependence on sample quality
PacBio / HiFi and SMRT Typically around 15–25 kb Highly accurate long reads, complex genes, indels, genome assembly ONT has an advantage for ultra-long reads; instrument investment
Illumina / SBS Typically 2 × 150 bp SNPs, large sample cohorts, RNA expression, clinical research Long repeats, structural variants, haplotype resolution
MGI / Complete Genomics / DNBSEQ Primarily short reads around 150 bp Large-scale whole-genome sequencing, cost efficiency Limited ability to resolve long-range genomic structures
Element Biosciences / ABC Up to 2 × 300 bp High-accuracy short reads, flexible operation, multiomics Cannot directly resolve very long genomic structures
Ultima Genomics / Flow-based SBS Short-read sequencing Very high throughput, low costs, rare SNVs Challenges involving estimation of repeated nucleotide counts
Roche / SBX and AXELIOS 1 Up to approximately 1,500 bp Rapid sequencing, high throughput Not ultra-long-read sequencing; limited independent validation history

The read lengths shown represent typical operating conditions or published specifications rather than absolute physical limits. [3][4][5][6][7][8][9]

It is also important not to rank these platforms by a single measure of "accuracy."

Single-read accuracy, consensus accuracy, SNP detection accuracy, and structural variant sensitivity measure different aspects of sequencing performance.

A technology that performs exceptionally well in one application may be less suitable for another.

4. The Most Direct Competitor: PacBio

4.1. How PacBio Sequencing Works

Pacific Biosciences (PacBio) is one of ONT's most direct competitors because both companies specialize in sequencing long DNA molecules.

Its representative platforms include Revio and Vega.

PacBio uses a technology called Single Molecule Real-Time Sequencing (SMRT Sequencing).

Rather than measuring changes in electrical current, PacBio observes fluorescence signals generated as DNA polymerase synthesizes DNA.

An important characteristic of the technology is that DNA molecules are circularized, allowing the same original DNA molecule to be read repeatedly.

By combining information from multiple passes, PacBio can obtain exceptionally accurate sequences from individual molecules.

These are known as HiFi reads (High-Fidelity reads). [3]

4.2. PacBio's Strengths

PacBio offers several major advantages:

  • Single-read accuracy of approximately 99.9% or higher.
  • Highly accurate sequencing of DNA molecules typically around 15–25 kb long.
  • Strong performance in detecting SNPs and small insertions or deletions (indels).
  • The ability to distinguish highly similar genes and pseudogenes.
  • Effective analysis of repetitive sequences and complex genomic regions.
  • Full-length RNA transcript analysis.

PacBio is particularly suitable when both substantial read length and very high nucleotide-level accuracy are required.

In 2026, improvements involving SPRQ-Nx chemistry and DeepConsensus have further increased sequencing accuracy and improved the economics of large-scale Revio sequencing and smaller-scale in-house Vega applications. [10][11]

PacBio has also expanded its capabilities for detecting DNA modifications, including 5mC, 5hmC, and 6mA. [10]

4.3. PacBio's Limitations

One of PacBio's fundamental trade-offs arises from its strategy of repeatedly reading the same DNA molecule.

There is an inherent relationship between molecule length, the number of passes, sequencing accuracy, and throughput.

ONT can read DNA molecules exceeding 100 kb and, under certain conditions, more than 1 Mb.

PacBio HiFi sequencing generally does not operate at these ultra-long read lengths.

For example, when a researcher needs a single molecule to span an enormous repetitive region and connect unique sequences on either side, ONT ultra-long reads offer an important advantage.

Instrument investment is another factor that laboratories and businesses must consider.

4.4. PacBio and ONT Are Both Competitors and Complementary Technologies

The two companies address sequencing accuracy through fundamentally different approaches.

ONT interprets electrical signals using increasingly sophisticated computational models.

PacBio improves accuracy by observing the same DNA molecule multiple times and integrating the results.

In simplified terms:

PacBio excels at reading relatively long DNA molecules with exceptionally high accuracy.

ONT excels at reading even longer DNA molecules directly.

These differences make the platforms competitors, but also highly complementary technologies.

Using both can enable analyses that would be difficult to accomplish with either platform alone.

5. The Largest Market Competitor: Illumina

5.1. How Illumina Sequencing Works

Illumina specializes in sequencing enormous numbers of relatively short DNA fragments with high accuracy.

Its major platforms include NovaSeq X, NextSeq, and MiSeq i100.

The underlying technology is called Sequencing by Synthesis (SBS).

DNA is fragmented into short pieces and amplified into clusters.

Fluorescently labeled nucleotides are incorporated sequentially, and their identities are determined through optical detection.

The latest generations use XLEAP-SBS chemistry. [4]

5.2. Illumina's Strengths

Illumina's advantages include:

  • Cost-efficient sequencing of large numbers of samples.
  • High-accuracy detection of SNPs and small indels.
  • Gene expression analysis using RNA-seq.
  • Targeted sequencing, including cancer gene panels.
  • Extensive software ecosystems, standardized procedures, and validated workflows.

For research involving hundreds, thousands, or tens of thousands of samples, short-read sequencing platforms such as Illumina remain extremely powerful.

They are particularly suitable for applications requiring consistent data generation across large sample cohorts.

5.3. Illumina's Limitations

Illumina's fundamental limitation is its relatively short read length.

This creates structural challenges when analyzing complex genomic regions.

For example, imagine that a genome contains five almost identical copies of a particular sequence.

A 150 bp fragment may not contain enough unique information to identify which copy it originated from.

Long-read technologies such as ONT and PacBio can often resolve this problem by reading from one distinctive sequence through the repeated region to another distinctive sequence.

In addition, conventional Illumina sequencing does not directly detect DNA chemical modifications through their effects on electrical current.

DNA methylation analysis therefore generally requires additional sample preparation or specialized methods.

6. Emerging Competitors in Short-Read Sequencing

Several companies are challenging Illumina by developing alternative approaches to short-read sequencing.

MGI, Element Biosciences, and Ultima Genomics are particularly important examples.

Their strategies emphasize sequencing accuracy, operating flexibility, throughput, and cost efficiency.

6.1. MGI / Complete Genomics: DNBSEQ

MGI's DNBSEQ technology begins by circularizing DNA molecules.

Rolling Circle Amplification is then used to generate DNA Nanoballs.

These nanoballs are arranged in high-density arrays, and DNA sequences are determined through fluorescence-based detection.

Representative instruments include DNBSEQ-T7. [5]

The technology's advantages include:

  • Cost competitiveness for large-scale genomic analysis.
  • Methods designed to reduce amplification-related duplication and bias.
  • Measures intended to minimize index hopping.
  • Suitability for large-scale whole-genome sequencing.

However, DNBSEQ shares the fundamental read-length limitations of other short-read technologies.

Long repetitive regions and complex structural variants can therefore require complementary long-read sequencing.

6.2. Element Biosciences: AVITI and AVITI24

Element Biosciences uses a technology called Avidite Base Chemistry (ABC).

Its major platforms include AVITI and AVITI24.

The technology is designed to deliver highly accurate short-read sequencing while supporting flexible operation, including laboratories processing relatively small numbers of samples.

It supports read lengths of up to 2 × 300 bp. [6]

AVITI24 also extends beyond conventional DNA sequencing.

The platform incorporates approaches for analyzing DNA, RNA, cellular morphology, proteins, and other aspects of cellular biology.

This reflects a broader evolution toward multiomics, in which several types of biological information are integrated to characterize cellular states and functions.

However, Element's short-read technology cannot directly provide the extremely long DNA molecules available through ONT or PacBio.

6.3. Ultima Genomics: UG 100

Ultima Genomics employs Flow-based SBS technology.

Its representative instrument is UG 100.

In this approach, different nucleotide types are introduced sequentially, and the system estimates how many nucleotides are incorporated during each flow.

The company emphasizes extremely high-throughput sequencing at low cost.

It is also developing ppmSeq, a method intended to detect very rare single-nucleotide variants (SNVs). [7]

One technical challenge of flow-based sequencing is accurately determining the number of identical nucleotides in consecutive runs.

For this reason, performance in applications such as indel detection requires careful validation.

These companies compete primarily with Illumina.

In relation to ONT, however, their technologies often serve complementary roles because short-read and long-read sequencing address different analytical problems.

7. A Particularly Interesting New Technology: Roche Sequencing by Expansion (SBX)

7.1. Nanopore Sequencing with a Fundamentally Different Approach

One of the most interesting developments in 2026 is Roche's Sequencing by Expansion (SBX) technology.

Roche announced the technology in 2025 and announced the global commercial launch of its AXELIOS 1 platform on June 29, 2026.

The research-use platform was also launched in Japan on July 1, 2026. [8][9]

Like ONT, Roche SBX uses biological nanopores and electrical signals.

However, the underlying approach is substantially different.

ONT passes native DNA molecules through nanopores, while Roche converts DNA sequence information into specially engineered molecules before sequencing.

7.2. Xpandomers: Expanding DNA Information by Approximately 50 Times

In Roche's SBX system, DNA sequence information is converted into an engineered molecule called an Xpandomer.

The Xpandomer is designed to be approximately 50 times longer than the original DNA.

Its structure is engineered to generate electrical signals that are easier to distinguish for different nucleotide identities.

The Xpandomer is then passed through a nanopore, and the electrical signals are interpreted to recover the original DNA sequence. [8]

In other words, rather than attempting to measure difficult electrical signals from natural DNA directly, Roche transforms the information into a molecule specifically designed for easier measurement.

This is conceptually different from ONT's approach.

ONT relies on increasingly sophisticated machine-learning algorithms to interpret complex signals generated by natural nucleic acids.

Roche attempts to simplify the signals themselves by engineering the molecules that produce them.

From the perspective of measurement science and molecular engineering, this is an especially interesting development.

7.3. Roche SBX's Strengths

Roche SBX offers several potential advantages:

  • Molecular engineering designed to reduce interference between nucleotide signals.
  • High throughput through the parallel use of millions of nanopores.
  • Near-real-time electrical readout.
  • Support for reads longer than those typically generated by conventional short-read platforms.
  • A system designed for rapid whole-genome sequencing.

According to Roche, AXELIOS 1 supports research workflows capable of completing human whole-genome sequencing within the same day.

Under suitable sample and library preparation conditions, reads up to approximately 1,500 bp can be obtained. [8]

Published specifications in Japan indicate that the SBX-Duplex approach can generate approximately 1.8–2.7 Tb of data in a four-hour sequencing run, with average base accuracy exceeding Q38. [9]

These figures represent manufacturer-reported performance under specified conditions and should not be interpreted as guaranteed results for every application.

7.4. Roche SBX's Limitations

SBX also has limitations that distinguish it from ONT.

First, it is not designed for ultra-long DNA sequencing at the 100 kb scale or beyond.

Therefore, it does not directly compete with ONT in applications that require a single DNA molecule to span enormous repetitive regions.

Second, SBX requires conversion of DNA information into engineered molecules.

This makes the quality and efficiency of the conversion process important factors in the overall workflow.

Finally, because the platform entered the market in 2026, it has a shorter history of independent validation across laboratories and difficult genomic applications than more established sequencing technologies.

Roche SBX can therefore be viewed as an extension of the fundamental idea of nanopore-based electrical sequencing, but with an important additional step: engineering the molecules being measured.

8. Which Sequencing Technology Is Best for Each Application?

The most suitable sequencing technology depends heavily on the analytical objective.

The following table provides a broad evaluation based on the physical principles and product capabilities of the major technologies as of 2026.

Legend:

  • ◎ Particularly well suited
  • ○ Well suited
  • △ Suitable under certain conditions
  • — Not a primary application
Application ONT PacBio Short-Read SBS Roche SBX
SNPs and small variants ○ ◎ ◎ ○
Large structural variants ◎ ◎ △ △
Very large repetitive regions ◎ ○ — △
High-quality de novo genome assembly ◎ ◎ △ △
DNA methylation analysis ◎ ◎ △ ○
Direct RNA sequencing ◎ — — —
Cost efficiency for large sample cohorts ○ ○ ◎ ◎
Rapid analysis in field settings ◎ △ △ ○

The short-read SBS category broadly includes Illumina, MGI, Element, and Ultima, although actual performance varies by platform.

Some assessments, including those related to Roche methylation workflows, also depend on the validation status of newer applications.

The central lesson is that no single sequencing technology dominates every application.

The most appropriate platform depends on whether the objective is to obtain long molecular reads, achieve exceptionally high accuracy, analyze large sample cohorts at low cost, or detect additional molecular information.

It is also important to distinguish the ability to detect chemical modifications from the accuracy and quantitative reliability of such detection.

Two platforms capable of detecting DNA methylation do not necessarily provide equivalent performance across all sequences and sample types.

9. What Can Be Achieved by Combining Sequencing Technologies?

In many applications, combining sequencing technologies can produce better results than relying on a single platform.

Four representative examples illustrate why.

9.1. Building a Highly Complete Genome Assembly for a New Species

Recommended combination: ONT Ultra-Long Reads + PacBio HiFi

ONT ultra-long reads can be used to connect large repetitive regions and reconstruct chromosome-scale structures.

PacBio HiFi reads can then provide highly accurate nucleotide information to support sequence reconstruction.

This combination is particularly useful for:

  • Plant genome sequencing.
  • Genomes containing extensive repetitive regions.
  • Complex chromosomal structures.
  • Telomere-to-telomere (T2T) genome assembly.

For projects prioritizing maximum genome assembly quality, this can be a powerful strategy.

However, using both technologies increases costs and sample requirements.

If sufficiently high-quality results can be obtained using ONT or PacBio alone, combining both platforms may not be necessary.

9.2. Comprehensive Analysis of Cancer-Associated Genetic Variants

Recommended combination: Illumina or Other Short-Read SBS + ONT or PacBio

Short-read sequencing can be used to analyze large numbers of samples for SNPs, indels, and low-frequency variants.

Long-read sequencing can then investigate regions that are difficult to resolve with short reads, including gene fusions, large deletions, and complex structural variants.

Rather than sequencing every sample using long-read technology from the beginning, a two-stage strategy may be more economical.

First, screen a large number of samples using short-read sequencing.

Then apply long-read sequencing to selected samples or genomic regions that require additional resolution.

However, adding long-read sequencing does not automatically improve the detection of extremely rare tumor variants.

Approaches involving Unique Molecular Identifiers (UMIs), duplex consensus, and adequate molecular sampling remain important.

9.3. Analyzing RNA Expression and Alternative Splicing

Recommended combination: Illumina / Element + ONT Direct RNA or PacBio Iso-Seq

Short-read RNA-seq is an efficient method for measuring gene expression across large numbers of genes and samples.

Long-read technologies such as ONT and PacBio provide additional information about transcript structure.

For example, they can help determine which exons are included in a particular RNA molecule and distinguish different transcript isoforms.

ONT Direct RNA Sequencing is particularly distinctive because it can read native RNA molecules without reverse transcription or PCR amplification.

This makes it useful not only for RNA sequence analysis but also for research into RNA chemical modifications.

Such capabilities support a transition from measuring gene expression levels alone toward investigating the structure and molecular state of individual RNA transcripts.

9.4. Rapid Detection of Infectious Diseases and Unknown Pathogens

Recommended combination: ONT as the Primary Platform, with Short-Read Confirmation When Needed

ONT can begin analyzing DNA or RNA sequences as soon as they are generated.

This makes it attractive for:

  • Infectious disease analysis.
  • Pathogen identification.
  • Detection of antimicrobial resistance genes.
  • Field-based infectious disease surveillance.

ONT also offers Adaptive Sampling.

This approach allows unwanted DNA molecules to be rejected from nanopores during sequencing, increasing the proportion of data obtained from target molecules.

However, diagnostic sensitivity depends on factors such as nucleic acid extraction quality, the amount of host DNA, and pathogen abundance.

A fast sequencing platform does not automatically guarantee high sensitivity in clinical diagnostics.

10. Where Is DNA Sequencing Technology Heading?

Based on current technological progress, four major directions are likely to shape future competition.

10.1. Differences in Basic Sequencing Accuracy May Continue to Narrow

ONT has substantially improved its sequencing accuracy.

PacBio is also becoming increasingly accurate and cost-effective.

As a result, differentiating platforms solely through nucleotide-level sequencing accuracy may become more difficult.

Other performance characteristics may become increasingly important:

  • How many kilobases or megabases can be read in a single molecule?
  • Which genomic regions can be analyzed without gaps?
  • Which chemical modifications can be detected?

These capabilities cannot be adequately evaluated using single-read nucleotide accuracy alone.

10.2. Sequencing Will Expand Beyond DNA Sequences to Chemical Modifications and Molecular Structure

ONT can detect DNA methylation and other modifications alongside DNA sequence information.

PacBio has also expanded its methylation analysis capabilities.

In the future, competitive differentiation may increasingly depend on the ability to characterize not only the nucleotide sequence of DNA but also its epigenetic state.

This represents an evolution from determining what a genome contains toward investigating the molecular state in which that genome exists.

10.3. AI and Computational Algorithms Will Increasingly Determine Sequencing Performance

Technologies such as ONT Dorado and PacBio DeepConsensus transform raw measurement signals into usable sequence information.

Machine-learning algorithms therefore play a critical role in sequencing performance.

Even when data are generated using the same hardware, improved computational models may produce better sequencing results.

PacBio, for example, has continued improving HiFi sequencing accuracy through advances involving DeepConsensus. [11]

However, computational algorithms cannot eliminate every physical measurement limitation or sample preparation problem.

Both molecular engineering and computational science remain essential.

10.4. Nanopore Technology May Expand Beyond DNA and RNA

A particularly interesting long-term opportunity involves applying nanopore technology to molecules other than nucleic acids.

The underlying principle of ONT sequencing is not fundamentally dependent on recognizing DNA itself.

Rather, it involves measuring how molecules passing through nanopores influence ionic currents.

In principle, this approach could be extended to peptides, proteins, and other molecules.

Research on nanopore-based protein identification and peptide analysis is already progressing.

However, as of 2026, nanopore-based protein sequencing has not yet reached the maturity of DNA sequencing as a general-purpose technology for determining complete protein sequences.

Over the longer term, it may become complementary to, or competitive with, mass spectrometry-based proteomics.

11. How Should Entrepreneurs Choose a Sequencing Technology?

The technological differences discussed above have direct implications for entrepreneurs building businesses around genomic information.

11.1. When the Goal Is to Resolve Complex Genome Structures

ONT ultra-long sequencing combined with PacBio HiFi is a strong candidate.

The combination can provide both exceptionally long reads and highly accurate nucleotide sequences.

However, businesses should consider whether a single technology can achieve the required result at lower cost and with fewer sample requirements.

11.2. When the Goal Is Large-Scale SNP or Gene Expression Analysis

Short-read technologies from Illumina, MGI, Element Biosciences, and Ultima Genomics are important options.

When processing thousands or tens of thousands of samples, cost efficiency per sample becomes particularly important.

11.3. When the Goal Is Flexible, Small-Scale, or Rapid Molecular Analysis

ONT is particularly attractive.

It supports ultra-long DNA sequencing, direct RNA analysis, DNA methylation detection, and rapid sequencing in field settings.

Its ability to analyze DNA and RNA molecules in relatively native forms is also a distinctive advantage.

11.4. When the Goal Is to Evaluate High-Speed, High-Throughput Next-Generation Technologies

Roche SBX deserves consideration.

Its strategy of converting DNA sequence information into engineered molecules is designed to improve nanopore signal quality and sequencing speed.

However, as a relatively new technology, its performance should be carefully validated for the intended business application.

11.5. Total Cost per Sample Matters More Than Cost per Gigabase

For a commercial sequencing business, comparing instrument prices or reagent costs alone is insufficient.

The more useful metric is:

The total cost per sample successfully analyzed at the required level of quality and accuracy.

This includes:

  • Instrument acquisition costs.
  • Library preparation expenses.
  • Sequencing reagents and consumables.
  • Experimental and analytical failure rates.
  • Computational infrastructure requirements.
  • Instrument utilization rates.
  • Labor costs for sequencing and data analysis.

Depending on the application, DNA or RNA storage stability and extraction quality may also be important.

The business objective is not simply to obtain nucleotide sequences at the lowest possible price.

It is to obtain the information required for a particular application, at the necessary level of accuracy, within the required timeframe, and at a commercially viable cost.

12. Understanding the Future Competitive Landscape

The current sequencing market can be understood in terms of four broad technological strategies.

Technology Primary Competitive Dimension Fundamental Approach
Oxford Nanopore Ultra-long reads and direct molecular measurement Direct analysis of very long DNA and RNA molecules, including sequence, structure, and chemical modifications
PacBio Highly accurate long reads High-accuracy sequencing of long DNA molecules to resolve complex genes and subtle variants
Illumina, MGI, Element, Ultima Scale and economics Efficient generation of high-quality sequencing data from large numbers of samples
Roche SBX Speed and signal engineering Conversion of DNA information into engineered molecules to improve nanopore signal quality and throughput

One particularly fascinating aspect of this competitive landscape is that ONT, PacBio, and Roche are approaching the same fundamental problem from different directions.

PacBio increases accuracy by repeatedly measuring the same molecule and combining the observations.

ONT improves accuracy by applying increasingly sophisticated computational analysis to electrical signals.

Roche redesigns the molecules generating the signals to make those signals easier to measure.

In other words, these companies are proposing different solutions based on physics, molecular engineering, and information science to the same challenge: accurately reading DNA sequences.

Over the next five to ten years, understanding the accuracy limits, physical constraints, and potential cost reductions of these approaches will be essential for evaluating the direction of sequencing technology.

For entrepreneurs seeking to build businesses using genomic information, the key opportunity may not lie in selecting a single sequencing technology.

It may instead lie in understanding how different technologies can be combined, what additional information they can generate, and how that information can be turned into valuable products and services.

The competitive landscape is evolving from a race to read DNA sequences as cheaply as possible toward a broader competition to extract diverse forms of information from DNA and RNA molecules.

The most important question may no longer be simply how cheaply we can read DNA, but how efficiently we can transform molecular information into real-world value.

That shift may ultimately be one of the most important developments in the future of DNA sequencing.

References

[1] Oxford Nanopore Technologies, "Continuous development and improvement," Corporate Technology Resources. https://nanoporetech.com/about/continuous-development-and-improvement

[2] Oxford Nanopore Technologies, "Nanopore sequencing accuracy," Corporate Technology Resources. https://nanoporetech.com/platform/accuracy

[3] Pacific Biosciences, "PacBio Revio | Long-read sequencing at scale," Product and Technology Resources. https://www.pacb.com/revio/

[4] Illumina, "Sequencing by Synthesis (SBS) Technology," Corporate Technology Resources. https://www.illumina.com/science/technology/next-generation-sequencing/sequencing-technology.html

[5] MGI, "DNBSEQ Technology," Corporate Technology Resources. https://us.global-mgitech.com/technologies/dnbseq-technology/

[6] Element Biosciences, "AVITI24: Next Gen Multiomic Platform," Product and Technology Resources. https://www.elementbiosciences.com/products/aviti24

[7] Ultima Genomics, "UG 100 Sequencing Platform," Product and Technology Resources. https://www.ultimagenomics.com/products/ug-100-sequencing-platform/

[8] Roche Diagnostics, "Roche announces the launch of AXELIOS 1, a transformative next-generation sequencing platform," Press Release, June 29, 2026. https://diagnostics.roche.com/us/en/news-listing/2026/axelios-1-next-generation-sequencing-platform.html

[9] Roche Diagnostics Japan, "Launch of AXELIOS 1, a Next-Generation Sequencer Powered by Proprietary SBX Technology: An Important Step Toward the Future of Genomic Medicine," Press Release, July 1, 2026. https://www.roche-diagnostics.jp/media/releases/2026-7-1

[10] Pacific Biosciences, "SPRQ-Nx on Vega raises the bar for benchtop HiFi sequencing," PacBio Blog, August 20, 2026. https://www.pacb.com/blog/sprq-nx-on-vega-raises-the-bar-for-benchtop-hifi-sequencing/

[11] Pacific Biosciences, "Improving HiFi sequencing accuracy with Google DeepConsensus and AlphaEvolve," PacBio Blog, April 30, 2026. https://www.pacb.com/blog/improving-hifi-sequencing-accuracy-with-google-deepconsensus-and-alphaevolve/