Digital & Molecular Forensics

The End of Traditional DNA Profiling?

How Long-Read DNA Sequencing (Oxford Nanopore & PacBio) Is Transforming Forensic Science

For three decades, short tandem repeat (STR) profiling read by capillary electrophoresis has been the backbone of forensic human identification. It still is. But a new class of sequencing technology — long-read, third-generation sequencing from Oxford Nanopore Technologies and Pacific Biosciences — is now being tested against exactly the cases that break traditional methods: skeletal remains, mass disaster victims, complex mixtures, and degraded evidence. This article examines what long-read sequencing actually is, what it can and cannot yet do in a forensic laboratory, and how close it really is to routine casework.

Reading time: 34–38 min Updated: July 2026 Author: Budding Forensic Expert Editorial Desk Category: Forensic Genetics / Emerging Technology

Table of Contents

1. A Case STR Profiling Could Not Close
2. The Evolution of DNA Profiling
3. Why Traditional STR Profiling Has Limits
4. What Is Long-Read DNA Sequencing?
5. Oxford Nanopore Technologies (ONT)
6. Pacific Biosciences (PacBio)
7. Oxford Nanopore vs PacBio
8. Forensic Applications
9. Why Long Reads Solve What Short Reads Cannot
10. Can Long Reads Replace STR Profiling?
11. Challenges to Routine Adoption
12. The Future of DNA Profiling
13. Recent Research (2023–2026)
14. Myth vs Reality
15. Frequently Asked Questions
16. Glossary
17. References

A Case STR Profiling Could Not Close

Somewhere in an evidence store, a set of skeletal remains has waited for identification for over a decade. Years of exposure degraded the DNA into fragments only a few hundred base pairs long. Every attempt at standard short tandem repeat (STR) profiling using capillary electrophoresis (CE) produced partial, unreliable, or drop-out-riddled profiles — not enough loci for a confident CODIS-style match, and nowhere near enough for a familial search. The case sits, unresolved, in the category every forensic geneticist knows well: technically unsolvable with the tools on hand.

This scenario is not rare. Disaster victim identification (DVI) teams, cold-case units, and missing-persons investigators across the world encounter it routinely — commingled remains, fire-damaged bone, decades-old bloodstains. Traditional STR-CE profiling was engineered for a different problem: relatively intact reference-quality DNA, a small, well-validated panel of loci, and speed. It excels at that problem. It struggles when the DNA itself is the obstacle.

Long-read DNA sequencing — sometimes called third-generation sequencing (TGS) — approaches the same evidence differently. Instead of reading DNA in short, uniform fragments and reassembling them statistically, platforms from Oxford Nanopore Technologies (ONT) and Pacific Biosciences (PacBio) read individual DNA molecules end-to-end, including the sequence context around and between forensic markers. Recent forensic genetics literature frames long-read sequencing as a promising complement to — not yet a replacement for — established STR methods, one that still requires validation, standardization, and cost reduction before it enters routine casework.^[1] This article lays out, without exaggeration, exactly where that promise currently stands.

Important

Nothing in this article should be read as suggesting long-read sequencing is currently validated for routine evidentiary casework in the way CE-based STR profiling is. Where research is early-stage or unvalidated, this article says so explicitly.

It is worth being precise about what "the end of traditional DNA profiling" would actually require. It is not enough for a new technology to work in a research laboratory once. Forensic methods that end up in a courtroom must clear a much higher bar: repeatable accuracy across thousands of samples, transparent error rates that a defence expert can cross-examine, standard operating procedures that any accredited laboratory can follow identically, and — critically — acceptance by the scientific community and the judiciary that the method is reliable enough to help decide guilt or innocence, custody, or the identity of the dead. STR-CE profiling took the better part of two decades to earn that status after PCR made it feasible in the first place. Long-read sequencing is, by comparison, only a few years into that same journey for forensic-specific use, even though the underlying technology itself is now mature enough to sequence entire human genomes routinely in research and clinical settings.

This article is written for exactly the audience that will decide how that journey unfolds: forensic science students who will staff tomorrow's DNA laboratories, researchers designing the validation studies that will eventually justify (or rule out) routine use, and practitioners — police officers, crime scene investigators, digital and biological forensic analysts, lawyers, and judges — who will need to understand the technology well enough to evaluate it critically rather than either dismissing it prematurely or over-trusting it because it sounds cutting-edge.

The Evolution of DNA Profiling

To understand why long-read sequencing matters, it helps to see forensic DNA analysis as a continuous search for more information from smaller, more damaged samples — a fifty-year trajectory of shrinking sample requirements and growing genetic resolution.

Restriction Fragment Length Polymorphism (RFLP)

The earliest forensic DNA typing method, RFLP, cut genomic DNA with restriction enzymes and compared fragment lengths on a gel. It was genuinely discriminating but required large amounts of high-quality DNA and took weeks — impractical for casework involving trace or degraded biological material. RFLP's forensic debut in the mid-1980s was nonetheless a watershed moment: for the first time, biological evidence at a crime scene could, in principle, be tied to one individual out of an entire population with a statistically defensible probability, rather than merely excluded from a small set of suspects using blood typing or serology.

RFLP's practical weaknesses, however, were severe by modern standards. A usable sample often needed to be roughly the size of a small bloodstain — quantities rarely available from trace evidence, weapons, or touch surfaces — and any degradation of the DNA (from heat, moisture, or time) destroyed the large fragments the method depended on. Forensic laboratories needed a way to work with far smaller, far more damaged samples, and to get answers in days rather than weeks.

PCR and STR-CE Profiling

The introduction of the polymerase chain reaction (PCR) in the late 1980s changed everything: minute amounts of DNA could now be amplified millions of times. Combined with short tandem repeats — short, highly variable repeat motifs scattered across the genome — and capillary electrophoresis for fragment sizing, this became the STR-CE system that underlies virtually every national DNA database today, including CODIS in the United States and comparable systems used by Indian forensic science laboratories and the NCRB.

Y-STR, X-STR, and Mitochondrial DNA

Lineage markers extended profiling to cases where autosomal STRs fail: Y-STR for male lineage tracing in sexual assault mixtures, X-STR for certain kinship configurations, and mitochondrial DNA (mtDNA) for severely degraded remains such as old skeletal material, since mtDNA exists in far higher copy number per cell than nuclear DNA.

SNP Typing and Next-Generation Sequencing (NGS)

Single nucleotide polymorphism (SNP) panels improved analysis of highly degraded samples (SNP amplicons can be designed shorter than STR amplicons) and enabled forensic DNA phenotyping — predicting externally visible characteristics from DNA. Massively parallel sequencing (MPS/NGS), using short-read platforms such as Illumina and Ion Torrent, then allowed many STR, SNP, and lineage markers to be sequenced simultaneously, revealing sequence-level variation within STR alleles that CE's length-only measurement had always masked.

This era also introduced microhaplotypes — clusters of closely linked SNPs inherited together as a block — as a forensic marker class distinct from both classical STRs and single, isolated SNPs. Microhaplotypes are attractive for exactly the samples that give STR-CE the most trouble: because they can be designed as short amplicons, they tend to survive degradation better than longer STR loci, and because they carry several linked variants at once, they can offer greater individual discriminatory power per marker than a single SNP while avoiding some of the stutter and slippage artefacts that complicate STR sequence interpretation. The rise of microhaplotypes is itself a preview of the broader shift long-read sequencing represents: forensic genetics moving from "how long is this fragment" toward "what is the full sequence, and what is linked to what."

Third-Generation, Long-Read Sequencing

The newest chapter is long-read sequencing, which reads individual DNA molecules — sometimes tens of thousands of bases long — directly and in real time. Named Method of the Year by Nature Methods in 2022 for its impact on genomics generally, long-read sequencing is now being explored specifically for the forensic problems that shorter reads and CE cannot resolve: repetitive regions, structural variation, and severely fragmented evidentiary DNA.

Why Traditional STR Profiling Has Limits

STR-CE profiling remains the gold standard for a reason: it is fast, inexpensive, extensively validated, court-tested, and supported by enormous reference databases. Its limitations are well documented and specific, not general.

Where STR-CE Struggles

Highly degraded DNA (fire, decomposition, burial), complex multi-contributor mixtures, low-template ("touch") DNA, distinguishing isoalleles of identical length but different sequence, resolving distant or complex kinship, and detecting structural variants or methylation states relevant to age or tissue-of-origin.

Degradation and Fragment Size

CE-based STR kits require amplicons typically in the 100–450 base pair range. When environmental exposure fragments DNA below this threshold — common in burned remains, buried bone, or old bloodstains — amplification fails at the larger loci first, producing partial profiles with reduced statistical power.

Mixture Interpretation

When two or more individuals contribute DNA to a single stain, CE electropherograms show overlapping peaks that must be deconvoluted using probabilistic genotyping software. As contributor number rises or ratios become skewed, confidence intervals widen substantially, and courts increasingly scrutinize how such mixtures are interpreted.

The Isoallele Problem

Because CE measures only fragment length, two alleles of identical length but different internal repeat-motif sequence appear indistinguishable — even though they are genetically different. Sequencing-based methods (NGS and long-read) can resolve this "isoallele" ambiguity directly.

Low-Template and Touch DNA

Trace or "touch" DNA yields very little starting template, increasing the risk of allele drop-out, drop-in, and stochastic amplification artifacts — a persistent challenge regardless of sequencing platform, though it particularly limits CE-STR reliability at the picogram scale.

Table 1: Common Forensic Sample Challenges vs STR-CE Performance

Sample Type	Typical Issue	STR-CE Outcome
Burned/charred bone	Severe fragmentation, PCR inhibitors	Partial or no profile
Decomposed soft tissue	Microbial DNA contamination, degradation	Reduced locus recovery
Multi-contributor mixture (3+)	Overlapping peaks, allele stacking	Complex, sometimes contested interpretation
Touch/trace DNA	Low copy number template	Drop-out, drop-in artifacts
Old skeletal remains (DVI)	Nuclear DNA loss, mtDNA-dominant	Reliance on mtDNA/lineage markers only

What Is Long-Read DNA Sequencing?

Definition

Long-read DNA sequencing (third-generation sequencing, TGS) is a class of sequencing technology that reads individual, native DNA molecules continuously and in real time, producing reads that can span thousands to tens of thousands of bases — far longer than the 100–300 base reads typical of second-generation (short-read) sequencing.

Two commercial approaches dominate the field: nanopore sequencing (Oxford Nanopore Technologies), which measures changes in electrical current as a single DNA strand passes through a protein pore, and single-molecule real-time sequencing (Pacific Biosciences), which optically detects fluorescently labelled nucleotides as a polymerase synthesizes a new strand within a nanophotonic well. Both are single-molecule methods: unlike short-read NGS, they do not require clonal amplification of DNA fragments on a solid surface before sequencing.

Key Concepts

• Single-molecule sequencing: each DNA (or RNA) molecule is read individually, not as part of a clustered, amplified population.
• Continuous, real-time output: data streams as sequencing occurs, rather than requiring a full imaging cycle per base.
• Read length: reads regularly span 10–25 kilobases and, in specialised ultra-long protocols, over a megabase.
• Error correction / consensus sequencing: raw single-pass reads have historically been more error-prone than short reads; both platforms use strategies (repeated circular reads for PacBio HiFi; improved chemistry and basecalling models for ONT) to raise accuracy.
• Direct detection of DNA modifications: both platforms can detect base modifications such as 5-methylcytosine directly from native DNA, without bisulfite conversion — relevant to forensic age estimation and tissue identification research.

Did You Know?

Long-read sequencing was named 2022 Method of the Year by Nature Methods, reflecting its growing role in resolving genome regions and variant types that short-read technology cannot reliably capture.

Oxford Nanopore Technologies (ONT)

Oxford Nanopore's platforms — MinION, GridION, and PromethION — sequence DNA by threading a single strand through a biological nanopore embedded in a membrane. As bases pass through the pore, they disrupt an ionic current in a distinctive, base-dependent pattern. Software ("basecalling," typically now handled by neural network models) translates this current signal back into a nucleotide sequence.

Instrument Range

• MinION: a USB-powered, pocket-sized device — the platform most explored for portable and field forensic applications.
• GridION: a benchtop device running up to five flow cells for moderate-throughput laboratory use.
• PromethION: a high-throughput system for population-scale or large-panel sequencing.

Advantages Relevant to Forensic Work

• Genuine portability — sequencing has been demonstrated outside conventional laboratories, including in outbreak and field settings.
• Direct sequencing of native DNA (and RNA) without mandatory PCR amplification in some workflows, though forensic-amount samples still typically require PCR enrichment of target loci.
• Ultra-long reads useful for spanning repetitive or structurally complex genomic regions.
• Native detection of methylation, relevant to emerging forensic age-prediction and body-fluid identification research.
• Lower upfront instrument cost compared to some competing platforms, which has driven interest from smaller and resource-limited laboratories.

Limitations

• Historically higher per-read error rates than short-read or PacBio HiFi sequencing, though improvements in flow cell chemistry (e.g., R10-series pores) and basecalling models have substantially narrowed this gap.
• Forensic-specific analysis software and validated bioinformatics pipelines remain limited compared to established CE-STR software.
• Early proof-of-concept forensic STR studies using MinION reported that high sequencing error rates constrained the reliability of full STR panels, although later work using purpose-built analysis tools such as STRspy achieved substantially improved concordance with reference genotypes.^[2][3]

Research Update

A proof-of-principle NIJ-supported study using the STRspy analysis pipeline on ONT MinION data reported robust, reliable genotyping across a full panel of autosomal STR loci amplified with a standard multiplex, with high concordance to manufacturer-validated reference profiles — while noting this remains a research demonstration rather than an accredited forensic method.^[2]

Pacific Biosciences (PacBio)

PacBio's Single Molecule, Real-Time (SMRT) sequencing detects fluorescently labelled nucleotides as a DNA polymerase incorporates them into a growing strand, inside a nanophotonic structure called a zero-mode waveguide (ZMW) that restricts the observation volume enough to detect a single incorporation event in real time.

Circular Consensus Sequencing and HiFi Reads

PacBio's key accuracy innovation is Circular Consensus Sequencing (CCS): a DNA template is circularised with hairpin adapters, and the polymerase reads around it multiple times. Averaging these repeated passes into a single consensus read — a "HiFi read" — yields read lengths typically in the 15–25 kilobase range with per-base accuracy independently benchmarked at roughly 99.9% (Q30) or higher, competitive with short-read platforms while retaining long-range sequence context.

Applications Relevant to Forensic Genetics

• Resolving structural variants, tandem repeat expansions, and other complex genomic regions in a single accurate read.
• Direct 5-methylcytosine and other base-modification detection without bisulfite treatment, of interest for forensic epigenetic age and tissue-source estimation research.
• High per-base accuracy suitable for confident individual allele calling, important where courtroom scrutiny of error rates is intense.

Limitations

• Instruments and consumables have historically carried a higher cost profile than nanopore devices, and are not field-portable in the way MinION is.
• Input DNA quantity and quality requirements can be more demanding, which is a meaningful constraint for trace or highly degraded forensic evidence.
• As with ONT, dedicated, validated forensic bioinformatics pipelines and inter-laboratory standardisation are still in development.

Oxford Nanopore vs PacBio: A Forensic Comparison

Neither platform is categorically "better" — each carries a different trade-off profile, and the right choice depends on the forensic problem being solved.

Table 2: ONT vs PacBio — General and Forensic-Relevant Comparison

Attribute	Oxford Nanopore (ONT)	PacBio (HiFi/SMRT)
Read length	Kilobases to >1 Mb (ultra-long protocols)	Typically 15–25 kb (HiFi)
Per-base accuracy	Improved substantially with R10 chemistry; still generally trails HiFi	~99.9% (Q30) or higher via consensus (HiFi)
Instrument size / portability	Pocket-sized (MinION) to benchtop	Benchtop/lab-scale (e.g., Revio); not field-portable
Relative instrument cost	Lower entry cost	Higher entry cost
Throughput (per run)	Scales with device tier (MinION to PromethION)	High-throughput systems process large sample volumes at scale
Methylation detection	Native detection supported	Native detection supported (e.g., 5mC, 5hmC, 6mA)
Field/on-site deployment	Demonstrated in non-laboratory settings	Laboratory-based only
Forensic bioinformatics maturity	Emerging, research-stage pipelines (e.g., STRspy)	Emerging, research-stage; fewer forensic-specific tools published
Best-suited forensic niche (current evidence)	Field/rapid triage, portable DVI scenarios (research stage)	High-accuracy resolution of complex/degraded lab samples

Important

Figures above reflect vendor-published specifications and peer-reviewed benchmarking as of 2024–2026 and are subject to change as chemistry and basecalling models continue to improve. Neither platform currently has a forensic STR kit validated to the level of established CE-STR systems for accredited casework use.

Forensic Applications of Long-Read Sequencing

Human Identification and Skeletal Remains

Because long reads can capture entire STR loci along with their flanking sequence context in a single molecule, researchers have explored their use for typing degraded skeletal DNA where CE amplicons routinely fail. Early studies stress that useful evidentiary information can be recovered even from partial or lower-quality data, though full-panel reliability still varies by sample condition.

Disaster Victim Identification (DVI)

Mass fatality events — building collapses, transportation disasters, fires — generate large numbers of severely degraded, sometimes commingled remains that must be identified quickly against reference and kinship samples. Portable nanopore sequencing has been proposed as a way to bring identification capability closer to the field, though its routine use in accredited DVI casework remains an active research question rather than settled practice.

The appeal for DVI specifically is scale and speed rather than any single technical trick: mass-fatality responses can generate hundreds or thousands of samples in a short window, and the bottleneck is often reference-database matching and family reference-sample logistics as much as the sequencing itself. A field-deployable device that can generate a usable partial profile within hours, even before samples reach a central laboratory for confirmatory analysis, could in principle shorten the agonising wait that families of disaster victims currently experience — provided the underlying method is validated to a standard DVI commissions and courts will accept.

Burned Human Remains

Heat severely fragments and chemically damages DNA. Long-read platforms cannot restore destroyed DNA, but their ability to work with lower-quality, fragmented starting material and to extract sequence-level information from what remains is an active area of forensic research.

Kinship Analysis and Missing Persons

Complex kinship scenarios — distant relatives, small reference pools, or disputed paternity/maternity cases — benefit from the additional genetic information (SNPs, sequence-level allele differences, structural variants) that sequencing-based methods can reveal beyond simple STR length matching. Missing-persons investigations are particularly demanding on kinship statistics because the reference sample available is often a distant relative — a great-aunt, a second cousin — rather than a parent or sibling, and the statistical power of a standard CODIS-style STR panel drops sharply as the genetic distance between the unknown and the reference increases. Denser, sequence-level genetic data, of the kind long-read platforms can generate alongside their core forensic markers, is one of several approaches being explored to extend kinship inference further out along the family tree.

Ancient DNA

Ancient and historical DNA research, distinct from routine forensic casework but methodologically related, has used long-read approaches to assemble degraded, fragmented genetic material and resolve repetitive regions that short reads struggle to place correctly.

Mixture Interpretation

Long-read, single-molecule sequencing has the theoretical advantage of preserving haplotype-level (phased) information — which alleles co-occur on the same physical DNA molecule — a property that could, in principle, aid deconvolution of complex multi-contributor mixtures beyond what allele-frequency-based CE interpretation allows. This application remains at the research and proof-of-concept stage.

Structural Variants and Repetitive Regions

STR loci themselves, along with other repetitive or structurally complex genomic regions, are exactly the regions long reads are best suited to span accurately in a single molecule — a documented general strength of the technology across genomics, now being tested specifically against forensic marker panels.

Methylation-Based Age and Tissue Prediction

Because both ONT and PacBio can detect DNA methylation directly from native molecules, researchers are exploring long-read methylation data as a route to forensic age estimation and body-fluid/tissue source identification — an active, still-emerging research area, not a validated forensic test.

Forensic Phenotyping, Microbial and Wildlife Forensics, Environmental DNA

Beyond human identification, long-read sequencing is being investigated for externally visible characteristic prediction, microbial forensic source attribution (useful in bioterrorism and outbreak investigation), wildlife forensics (species and provenance identification in trafficking cases), and environmental DNA analysis relevant to crime scene reconstruction.

Case Study Context

Peer-reviewed forensic genetics literature increasingly discusses long-read sequencing case studies and pilot datasets involving degraded and mixture samples, generally framed as demonstrating feasibility and evidential potential rather than as validated, court-ready casework methods.^[1]

Why Long Reads Solve Problems Short Reads and CE Cannot

The core advantage of long-read sequencing is contextual: a single read can span an entire region of interest — including repeats, flanking sequence, and nearby variants — without needing to be statistically reassembled from fragments. This matters for several persistently difficult forensic genomic features.

• Difficult genomic regions: segmental duplications, centromeric and telomeric regions, and other areas historically poorly resolved by short reads.
• Repetitive DNA: STR and other tandem repeat regions are read across in full, rather than inferred from fragment length alone.
• Structural variants: insertions, deletions, inversions, and copy-number changes larger than a short read can span are directly observable.
• Phasing and haplotype analysis: because a single molecule carries multiple variants together, which alleles are physically linked (in "cis" versus "trans") can be determined directly — valuable for complex kinship and mixture questions.
• Complex mixtures: phased, single-molecule reads carry more information per read about which alleles belong to which contributor than length-only CE data.
• Degraded DNA: while long reads cannot lengthen DNA that is physically fragmented, sequence-level analysis of whatever fragments remain can still extract more identifying information than length-based CE typing of the same fragments.

Quick Facts

• Long-read sequencing was named 2022 Method of the Year by Nature Methods.
• PacBio HiFi reads commonly reach 99.9%+ per-base accuracy via circular consensus sequencing.
• ONT's MinION is small enough for field and near-patient/near-scene deployment.
• Both platforms can detect DNA methylation directly, without bisulfite conversion.
• No long-read forensic STR kit is yet validated to CE-STR's accredited casework standard.

Can Long-Read Sequencing Replace STR Profiling?

Not currently, and not in the near term for routine casework — though the honest answer is more nuanced than a flat yes or no.

The Case For Eventual Change

Long-read sequencing resolves several genuine, well-documented STR-CE limitations: isoallele ambiguity, phasing, structural variant detection, and — potentially — improved recovery of information from degraded samples. As chemistry, accuracy, and cost continue to improve, sequencing-based forensic methods (short-read NGS first, long-read second) are widely expected to expand their footprint in forensic laboratories.

The Case For Caution

STR-CE profiling is backed by decades of validation studies, standardized commercial kits, enormous population allele-frequency databases, and an established body of case law affirming its admissibility. Long-read platforms currently lack: (1) forensic-specific kits validated to accredited standards, (2) standardized bioinformatics pipelines accepted across laboratories, (3) large-scale inter-laboratory concordance studies, (4) settled guidance from bodies such as SWGDAM or ENFSI on validation requirements, and (5) a track record of courtroom scrutiny comparable to CE-STR.

Expert Insight

Recent forensic genetics reviews consistently frame long-read sequencing as a complementary tool for the hardest cases — degraded remains, complex mixtures, difficult kinship — rather than a wholesale replacement for STR-CE, which remains fit for purpose in the large majority of routine casework.^[1]

Cost, Training, and Infrastructure

Even where the science is promising, practical adoption barriers remain significant: instrument and reagent costs, the need for bioinformatics expertise not traditionally part of forensic biology training, laboratory accreditation requirements (ISO 17025 and equivalents), and the multi-year timelines typically required to validate a new forensic method for court use.

Balanced conclusion: long-read sequencing is best understood today as an emerging complement for cases STR-CE cannot resolve — not an imminent replacement for the routine, high-throughput, well-validated STR-CE system that continues to underpin the vast majority of forensic DNA casework worldwide.

Challenges to Routine Forensic Implementation

Table 3: Key Barriers to Routine Forensic Adoption

Barrier	Why It Matters
Cost	Instrument, reagent, and per-sample costs remain higher than mature CE-STR workflows for routine volume casework.
Error rates	Even improved chemistries require rigorous, forensic-specific error characterization before evidentiary use.
Bioinformatics	Analysis pipelines are not yet standardized, validated, or widely available as accredited forensic software.
DNA quality/quantity requirements	Some long-read workflows still need more or higher-quality input DNA than ideal for trace evidence.
Standardisation	No universally adopted forensic long-read kit or protocol yet exists across laboratories.
Court validation	Admissibility requires an established, peer-reviewed track record that long-read methods have not yet accumulated.
Accreditation	ISO 17025 and equivalent frameworks require documented, reproducible validation before operational use.
Reference databases	Population allele-frequency databases built for CE-STR loci do not directly transfer to new sequence-level marker sets.

The Future of DNA Profiling

Predicting technology adoption timelines is always risky, and forensic science has historically been appropriately conservative about adopting new methods until validation catches up with capability. With that caveat, several trends visible in the wider genomics field are likely to shape how — and how quickly — long-read sequencing moves toward forensic practice over the coming decade.

• AI-assisted sequencing and basecalling: machine-learning basecallers are already improving ONT accuracy and are likely to keep narrowing the gap with short-read and HiFi accuracy.
• Portable sequencing at crime scenes: MinION-class devices point toward a future where preliminary sequencing occurs closer to the scene, though evidentiary-grade results will likely still require laboratory confirmation for the foreseeable future.
• Rapid sequencing workflows: shortening the time from sample to actionable data, valuable for time-critical missing-persons and disaster-response scenarios.
• Long-read STR typing kits: commercial development of forensic-specific long-read STR/SNP panels is a plausible next step as the underlying technology matures.
• Methylation biomarkers and epigenetics: age estimation, tissue/body-fluid identification, and even lifestyle-related epigenetic markers are active research fronts enabled by native methylation detection.
• Single-cell sequencing: extending sequencing resolution to individual cells could eventually assist mixture interpretation at a fundamentally different level of precision.
• Real-time DVI applications: combining portable sequencing with rapid bioinformatics could shorten identification timelines in future mass-fatality responses, pending validation.

Future Scope

Most experts frame the next five to ten years as a period of parallel use — STR-CE for routine, high-throughput casework, and long-read sequencing for a defined set of complex, high-value cases — rather than a rapid, wholesale technology switch.

Recent Research Breakthroughs (2023–2026)

Peer-reviewed forensic genetics literature published between 2023 and 2026 reflects a field moving from feasibility demonstrations toward more systematic evaluation:

• Comparative reviews of next-generation sequencing platforms for forensic science, published in 2024, have explicitly catalogued the advantages and disadvantages of short-read (Illumina, Ion Torrent) versus long-read (ONT, PacBio) approaches for forensic use, underscoring that each generation of technology answers different evidentiary questions.^[4]
• A 2025 forensic genetics review focused specifically on long-read sequencing potential highlighted persistent challenges in complex DNA mixture interpretation using CE-based probabilistic genotyping, and framed third-generation sequencing's real-time, single-molecule, long-read capabilities as a promising avenue for forensic research and, eventually, routine casework.^[1]
• Improvements in nanopore chemistry (R10-series pores) and neural-network basecalling models have been shown, in broader genomics validation studies, to push modal per-read accuracy and variant-calling F1 scores above 98% — a substantial improvement from earlier nanopore generations, though these studies are largely from clinical and research genomics rather than forensic-specific validation.^[5]
• Clinical long-read sequencing validation studies (2025) for broad genetic diagnosis illustrate the kind of rigorous, multi-sample concordance and error-characterization work that forensic long-read methods will similarly need before evidentiary adoption.^[5]

Research Update

As of the most recent published reviews, forensic long-read sequencing remains squarely in the research and pilot-validation stage. Authors in this space consistently distinguish established forensic practice (STR-CE) from emerging research (long-read sequencing), and call for continued validation and standardisation rather than reporting the technology as forensic-ready.^[1]

Myth vs Reality

Myth

"Long-read sequencing has already replaced STR profiling in forensic laboratories."

Reality

STR-CE profiling remains the standard, validated, court-accepted method for the overwhelming majority of forensic casework worldwide. Long-read sequencing is used in research and pilot studies, not as a routine casework replacement.

Myth

"Nanopore sequencing is always more accurate than STR-CE typing."

Reality

Nanopore accuracy has improved considerably but has historically trailed both short-read sequencing and PacBio HiFi consensus reads at the single-read level; accuracy depends heavily on chemistry version, coverage depth, and analysis pipeline.

Myth

"Long-read sequencing works perfectly on every forensic sample, however degraded."

Reality

No sequencing technology can recover genetic information from DNA that has been chemically destroyed. Long reads improve information extraction from surviving fragments; they do not repair fundamentally degraded DNA.

Myth

"Portable sequencers eliminate the need for forensic laboratories."

Reality

Portable devices such as MinION extend where preliminary sequencing can occur, but evidentiary-grade analysis, quality control, chain-of-custody, and accreditation still require laboratory infrastructure and oversight.

Frequently Asked Questions

What is long-read DNA sequencing in simple terms?

It is a sequencing method that reads an entire DNA molecule continuously, producing much longer stretches of sequence data per read than traditional short-read methods, which read only small fragments at a time.

Is long-read sequencing the same as next-generation sequencing (NGS)?

Long-read sequencing is sometimes considered a subset or a "third generation" beyond NGS. Traditional NGS (Illumina, Ion Torrent) is short-read; ONT and PacBio represent long-read, third-generation platforms.

What is the main difference between Oxford Nanopore and PacBio?

ONT reads DNA by detecting disruptions in electrical current as a strand passes through a nanopore. PacBio detects fluorescent light signals as a polymerase synthesizes DNA in a nanophotonic well. ONT devices are more portable; PacBio HiFi reads currently offer higher per-base accuracy.

Can long-read sequencing be used in court right now?

Not as a standard, validated forensic method comparable to STR-CE. It remains at the research and pilot-validation stage, without the accredited kits, standardized pipelines, and case-law history that CE-STR profiling has accumulated.

Will long-read sequencing eventually replace STR profiling entirely?

Most forensic genetics experts view this as unlikely in the near term. It is more probable that long-read sequencing becomes a complementary method for specific difficult cases, while STR-CE continues for routine casework.

What is a HiFi read?

A HiFi read is a PacBio consensus sequence generated by repeatedly sequencing a circularized DNA template and averaging the passes, producing long reads with very high per-base accuracy.

What is MinION?

MinION is Oxford Nanopore's smallest, USB-powered sequencing device, notable for its portability and use in field and resource-limited settings.

How does long-read sequencing help with degraded DNA?

It cannot restore physically destroyed DNA, but it can extract more identifying sequence-level information from whatever fragments remain, compared to length-only CE-STR typing of the same material.

What is an isoallele, and why does it matter?

An isoallele is an STR allele that has the same fragment length as another allele but a different internal repeat sequence. CE-STR, which measures only length, cannot distinguish them; sequencing-based methods can.

Can long-read sequencing help interpret complex DNA mixtures?

It has theoretical advantages because single-molecule reads preserve phased (linked) allele information, which could aid mixture deconvolution. This application is still largely at the research stage.

What is disaster victim identification (DVI), and how does long-read sequencing relate to it?

DVI is the process of identifying victims of mass-fatality events. Long-read sequencing, particularly portable nanopore devices, is being researched as a way to speed up identification of degraded or commingled remains, though it is not yet routine DVI practice.

Does long-read sequencing require less DNA than STR profiling?

Not necessarily. Some long-read workflows require comparable or greater quantities of higher-quality DNA than CE-STR, which is one of the practical barriers to forensic adoption, particularly for trace evidence.

What is methylation detection, and why is it relevant to forensics?

Methylation is a chemical modification of DNA that can vary with age and tissue type. Both ONT and PacBio can detect it directly from native DNA, which researchers are exploring for forensic age estimation and body-fluid identification.

Is Oxford Nanopore accurate enough for forensic casework?

Accuracy has improved significantly with newer chemistry and basecalling models, but ONT has not yet achieved the level of standardized, validated accuracy required for routine accredited forensic casework.

What is SMRT sequencing?

Single Molecule, Real-Time (SMRT) sequencing is PacBio's core technology, in which a polymerase synthesizes DNA inside a zero-mode waveguide while a camera detects each fluorescently labelled base in real time.

What is a zero-mode waveguide (ZMW)?

A ZMW is a nanophotonic well used in PacBio sequencing that restricts the observed volume enough to detect the fluorescent signal from a single nucleotide incorporation event.

Why can't short-read sequencing resolve repetitive DNA regions well?

Short reads are too brief to span an entire repetitive region uniquely; multiple short fragments from different parts of a repeat can look identical, making accurate reassembly difficult. Long reads span the whole region in one continuous read.

What organizations set forensic DNA validation standards?

Key bodies include SWGDAM (Scientific Working Group on DNA Analysis Methods) in the United States, ENFSI (European Network of Forensic Science Institutes), and ISFG (International Society for Forensic Genetics), alongside national accreditation bodies.

What does "third-generation sequencing" mean?

It refers to single-molecule, long-read sequencing platforms (ONT, PacBio) that came after first-generation Sanger sequencing and second-generation short-read NGS platforms.

Can long-read sequencing be used for kinship and missing-persons cases?

Researchers are exploring this application because sequencing-based methods can reveal more genetic information than STR length alone, which may help in distant or complex kinship scenarios, though it is not yet standard practice.

What is forensic DNA phenotyping?

It is the prediction of externally visible characteristics — such as eye or hair colour, or biogeographic ancestry — from DNA, typically using SNP panels, and increasingly explored using sequencing-based methods.

What is STRspy?

STRspy is a bioinformatics pipeline developed to analyze Oxford Nanopore sequencing data for forensic STR genotyping, reported in NIJ-supported research to achieve high concordance with reference STR profiles in proof-of-principle studies.

Are there portable DNA sequencers being used at crime scenes today?

Field and near-scene deployments of devices like MinION have been demonstrated in research and outbreak-response contexts, but routine, accredited use at active crime scenes for evidentiary purposes is not yet established practice.

What is the difference between short-read and long-read sequencing error profiles?

Short-read sequencing generally has lower per-base error rates but cannot span long or repetitive regions. Long-read sequencing spans much more sequence per read but has historically carried higher raw error rates, though consensus methods like PacBio HiFi substantially reduce this.

What is ancient DNA analysis, and how does it relate to forensic long-read sequencing?

Ancient DNA analysis studies historical or archaeological genetic material, often highly degraded. It shares technical challenges with forensic degraded-DNA work and has similarly benefited from long-read approaches to assembly and repeat resolution.

Does long-read sequencing help with wildlife or microbial forensics?

Yes — researchers are applying long-read sequencing to species identification in wildlife trafficking cases and to microbial source attribution in bioterrorism or outbreak investigations, alongside its human-identification applications.

What is the biggest barrier to forensic adoption of long-read sequencing?

No single barrier dominates; cost, standardisation, validated bioinformatics, accreditation requirements, and the absence of an established court track record all currently limit routine adoption.

Is long-read sequencing used anywhere in forensic casework today?

Its use remains largely confined to research studies, pilot validations, and academic publications rather than routine accredited casework, though this is an actively evolving area.

How long are typical long-read sequencing reads compared to STR amplicons?

CE-STR amplicons are typically 100–450 base pairs. Long-read sequencing reads commonly span 10,000–25,000 base pairs or more — vastly longer than a single STR locus requires, allowing multiple markers and their flanking context to be captured together.

What should students take away from the long-read vs STR-CE debate?

That forensic genetics evolves through careful, incremental validation rather than sudden replacement — new technologies expand what is possible for hard cases while established, validated methods continue to serve the majority of casework.

Glossary

Basecalling

The computational process of converting raw sequencer signal (electrical current for ONT, fluorescence for PacBio) into a DNA base sequence.

Capillary Electrophoresis (CE)

A method of separating DNA fragments by size using an electric field through a thin capillary; the detection method underlying standard STR profiling.

Circular Consensus Sequencing (CCS)

PacBio's method of repeatedly sequencing a circularized DNA template to generate a high-accuracy consensus "HiFi" read.

CODIS

Combined DNA Index System — the United States' national DNA database and its associated core STR loci panel.

Disaster Victim Identification (DVI)

The forensic process of identifying individuals killed in mass-fatality events.

ENFSI

European Network of Forensic Science Institutes, a body that develops shared forensic standards across Europe.

Isoallele

An STR allele identical in fragment length to another but differing in internal sequence composition.

ISFG

International Society for Forensic Genetics, a leading professional body for forensic genetic science.

Long-Read Sequencing (LRS)

Sequencing technology that reads DNA molecules continuously over long stretches, typically thousands to tens of thousands of bases.

Mitochondrial DNA (mtDNA)

DNA located in mitochondria, present in high copy number per cell, useful for severely degraded samples and maternal lineage tracing.

Nanopore Sequencing

A sequencing method that measures ionic current changes as a DNA strand passes through a nanoscale protein pore.

Phasing / Haplotype

Determining which genetic variants are physically located on the same DNA molecule (in "cis"), as opposed to simply knowing which variants are present.

Probabilistic Genotyping

Statistical software-based methods for interpreting complex or low-template DNA mixture profiles.

Short Tandem Repeat (STR)

A short DNA sequence motif repeated a variable number of times, widely used as the basis of forensic human identification.

Single Molecule, Real-Time (SMRT) Sequencing

PacBio's sequencing technology that observes DNA synthesis by a single polymerase molecule in real time.

SWGDAM

Scientific Working Group on DNA Analysis Methods, which issues U.S. forensic DNA validation guidelines.

Third-Generation Sequencing (TGS)

An alternative term for long-read, single-molecule sequencing technologies such as those from ONT and PacBio.

Zero-Mode Waveguide (ZMW)

A nanophotonic structure used in PacBio SMRT sequencing that isolates the optical signal from a single nucleotide incorporation event.

References

Peer-Reviewed Research Papers

• [1] Ferreira, M.R., Carratto, T.M.T., Frontanilla, T.S., Bonadio, R.S., Jain, M., de Oliveira, S.F., Castelli, E.C., Mendes-Junior, C.T. (2025). Advances in forensic genetics: Exploring the potential of long read sequencing. Forensic Science International: Genetics, 74, 103156. https://doi.org/10.1016/j.fsigen.2024.103156
• [2] National Institute of Justice. Accurate profiling of forensic autosomal STRs using the Oxford Nanopore Technologies MinION device. https://nij.ojp.gov/library/publications/accurate-profiling-forensic-autosomal-strs-using-oxford-nanopore-technologies
• [3] Cornelis, S., Willems, S., Van Neste, C., Tytgat, O., Weymaere, J., Vander Plaetsen, A.-S., Deforce, D., Van Nieuwerburgh, F. Forensic STR profiling using Oxford Nanopore Technologies' MinION sequencer. bioRxiv. https://www.biorxiv.org/content/10.1101/433151v1
• [4] Review of Six Different Next Generation Sequencing (NGS) Techniques for Forensic Science, including Advantages and Disadvantages. (2024). Current Forensic Science, 2(1), E120124225623. https://doi.org/10.2174/0126664844274727231218061037
• [5] Sen, S., Handler, H.P., Victorsen, A., et al. (2025). Validation of a comprehensive long-read sequencing platform for broad clinical genetic diagnosis. Frontiers in Genetics, 16, 1499456. https://doi.org/10.3389/fgene.2025.1499456

Official Manufacturer & Institutional Resources

• Oxford Nanopore Technologies. Forensic STR profiling using Oxford Nanopore Technologies' MinION. https://nanoporetech.com/resource-centre/forensic-str-profiling-using-oxford-nanopore-technologies-minion
• Pacific Biosciences. Long-Read Sequencing: Benefits & HiFi Accuracy. https://www.pacb.com/technology/long-read-sequencing/
• Pacific Biosciences. Understanding PacBio HiFi Sequencing. https://www.pacb.com/technology/hifi-sequencing/
• Pacific Biosciences. PacBio Revio System. https://www.pacb.com/revio/

Professional & Standards Bodies (for further reading)

• International Society for Forensic Genetics (ISFG). https://www.isfg.org
• European Network of Forensic Science Institutes (ENFSI). https://enfsi.eu
• Scientific Working Group on DNA Analysis Methods (SWGDAM). https://www.swgdam.org
• National Institute of Justice (NIJ). https://nij.ojp.gov

A Note on Sources

This article draws on peer-reviewed forensic genetics literature and official manufacturer technical documentation current as of 2024–2026. Readers requiring citation for academic or courtroom use should consult the original papers directly via the DOIs and URLs above rather than this summary.

Key Takeaways

• STR-CE profiling remains the validated, court-accepted standard for the vast majority of forensic DNA casework.
• Long-read sequencing (ONT and PacBio) offers real, documented advantages for degraded DNA, complex mixtures, kinship, and structurally complex regions — but remains at the research and pilot-validation stage forensically.
• ONT offers portability and ultra-long reads; PacBio HiFi offers higher single-read accuracy — neither is universally superior.
• Cost, standardisation, validated bioinformatics, and courtroom track record are the principal barriers to routine adoption.
• The most likely near-term future is parallel use: STR-CE for routine casework, long-read sequencing for the hardest cases.

Reader question: If long-read sequencing becomes routine in your country's forensic laboratories, should courts require a different evidentiary standard than the one applied to STR-CE profiling? Share your view.

Enjoyed this article? Share it with fellow forensic professionals and explore more in-depth forensic science content on Budding Forensic Expert.

Explore More on Budding Forensic Expert →

Written & reviewed by the Budding Forensic Expert Editorial Desk Category: Forensic Genetics / Emerging Technology · Last Updated: July 2026