DNA Sequencing Methods — Sanger, NGS, Long-Read, and What’s Next (2026)

DNA sequencing methods determine the order of the four bases — adenine (A), thymine (T), guanine (G), and cytosine (C) — in a DNA molecule. In 2026, four sequencing approaches dominate clinical and research practice: Sanger sequencing for targeted, single-gene work; short-read next-generation sequencing (NGS, e.g., Illumina) for whole-genome and whole-exome studies at scale; long-read sequencing (PacBio HiFi, Oxford Nanopore) for structural variants and complex regions; and emerging single-cell and spatial methods that profile sequence variation within individual cells. Each method trades off accuracy, read length, throughput, and cost — and the right choice depends entirely on what you’re trying to learn.

This guide explains how each method works, when to use it, what it costs in 2026, and how DNA sequencing data flows into clinical and research workflows.

The four DNA sequencing methods in clinical and research use today

1. Sanger sequencing — the gold standard, still

Sanger sequencing, developed by Frederick Sanger in 1977, was the method that produced the first human genome reference in 2003. In 2026 it’s still the gold standard for accuracy and remains in routine clinical and research use for specific applications.

How it works. A DNA strand is replicated in vitro using a polymerase and a mix of normal deoxynucleotides (dNTPs) plus a small fraction of fluorescently-labelled chain-terminating dideoxynucleotides (ddNTPs). Each ddNTP terminates replication at a different base. The fragments are separated by capillary electrophoresis, and a laser reads the fluorescent label at the end of each fragment to determine which base terminated it. The result is a chromatogram showing the sequence.

When to use Sanger in 2026:
– Confirming variants identified by NGS before clinical action — this is still the standard of care in many molecular diagnostics labs
– Sanger sequencing of single genes for known disease-causing variants (e.g., CFTR for cystic fibrosis confirmation)
– Verifying plasmid constructs in molecular biology workflows
– Forensic DNA analysis for STR (short tandem repeat) profiling
– Educational and small-scale research labs where NGS throughput is unnecessary

Cost: Outsourced Sanger sequencing in 2026 runs $5–$20 per reaction at standard CROs (GENEWIZ, Eurofins). Premium and specialty services run higher.

2. Short-read next-generation sequencing (NGS) — the workhorse

Short-read NGS, dominated by Illumina’s sequencing-by-synthesis chemistry, is the default approach for whole-genome (WGS) and whole-exome (WES) sequencing in clinical and research labs. The Broad Institute’s “$5 genome” target — driven by NovaSeq X high-throughput platforms — is the production reality at large genome centers in 2026.

How it works. DNA is fragmented into ~300 bp pieces, ligated to adapters with sample-specific barcodes, amplified by PCR, and loaded onto a flow cell. On the flow cell, each fragment is amplified into a clonal cluster, then sequenced by cycling through fluorescently-labelled reversible-terminator nucleotides. After each cycle, a camera images the entire flow cell to determine which base was incorporated at each cluster. Hundreds of millions to billions of clusters are sequenced in parallel.

Read lengths and platforms (2026):

When to use short-read NGS in 2026:
– Whole-genome sequencing (WGS) at 30× coverage for population studies, rare disease, oncology — the dominant approach for national programs like the UK Biobank, All of Us, and Genomics England
– Whole-exome sequencing (WES) for clinical diagnostics in rare-disease and oncology
– Targeted gene panels for clinical genomics (e.g., hereditary cancer, cardiology, pharmacogenomics)
– RNA-seq for transcriptomic profiling
– ChIP-seq, ATAC-seq, Hi-C for epigenomic and 3D genome analysis
– Microbiome / metagenomic sequencing

The 2026 cost reality: Production WGS at major sequencing centers (Broad, Wellcome Sanger, BGI) is now in the $200–$600 per genome range at 30× coverage. Smaller cores and CROs typically charge $600–$1,500. The “$1,000 genome” target from 2014 is well behind us — but downstream analysis (variant calling, annotation, interpretation) and storage costs now exceed the raw sequencing cost.

3. Long-read sequencing — solving the structural-variant problem

Short-read NGS is excellent at calling single-nucleotide variants (SNVs) and small indels but struggles with structural variants (large insertions, deletions, inversions, translocations), repetitive regions, and complex regions of the genome. Long-read sequencing solves these problems by producing reads of 10–100+ kilobases (PacBio HiFi) or even multi-megabase reads (Oxford Nanopore).

PacBio HiFi (HiFi = High Fidelity). Uses single-molecule real-time (SMRT) sequencing of circular consensus reads. Each DNA fragment is sequenced multiple times in a circle, then the reads are consensus-called to produce a single accurate long read. Typical accuracy: 99.9% at 15–25 kb read length. Production platforms: Revio, Sequel IIe.

Oxford Nanopore (ONT). Drives a single-stranded DNA molecule through a protein nanopore embedded in an electrically-charged membrane. As each base passes through the pore, it changes the ionic current in a base-specific way. The current signal is interpreted by a neural network basecaller. Read lengths can exceed 4 Mb. Accuracy at Q20+ chemistry runs 99%; raw single-pass reads are 95–99% depending on chemistry. Production platforms: MinION, GridION, PromethION 2 Solo, PromethION 48.

When to use long-read sequencing in 2026:
– Structural variant detection in rare disease, cancer, and population genomics
– Phasing — assigning variants to maternal vs paternal haplotypes
– Repetitive regions the short-read methods miss (centromeres, telomeres, ribosomal repeats, tandem repeat expansions like in Huntington’s disease)
– De novo genome assembly for non-model organisms
– Direct methylation detection without bisulfite conversion (both PacBio and ONT detect 5mC, 5hmC, 6mA natively)
– Real-time pathogen detection (ONT) — sequencing reads come off the platform in seconds, enabling clinical metagenomics
– Long-read RNA-seq for full-length transcript identification

Cost: Long-read WGS in 2026 runs $700–$1,400 per genome at clinically-relevant coverage (~25× HiFi or 30× ONT Q20+). The cost premium over short-read NGS is shrinking as platforms scale.

4. Single-cell and spatial sequencing — the new frontiers

Most DNA and RNA sequencing measures average signal across millions of cells. Single-cell sequencing measures signal per cell, revealing heterogeneity that bulk sequencing erases — critical for tumor biology, immune profiling, developmental biology, and cell-type discovery.

Single-cell RNA sequencing (scRNA-seq) captures the transcriptome of individual cells, typically by encapsulating each cell in a droplet with a barcoded bead (10x Genomics Chromium platform). Each transcript gets tagged with a cell-specific barcode before sequencing.

Single-cell DNA sequencing profiles genomic variation within individual cells — used for clonal evolution in cancer, somatic mutation profiling, and trisomy detection in pre-implantation embryos.

Spatial transcriptomics preserves the anatomical context of where each gene-expression measurement comes from in the tissue. Platforms in 2026: 10x Visium HD, NanoString CosMx, Vizgen MERSCOPE, Stereo-seq.

When to use single-cell or spatial sequencing in 2026:
– Tumor heterogeneity and resistance-clone tracking in oncology
– Immune cell profiling in immunotherapy response prediction
– Developmental biology and cell-fate trajectory mapping
– Drug-target validation — identifying which cell populations express the target
– Spatial tumor microenvironment characterization for solid tumors

Cost: Single-cell experiments in 2026 are still relatively expensive — a typical 10x Chromium scRNA-seq study capturing ~10,000 cells from one sample runs $2,000–$5,000 in reagents alone, plus sequencing costs. Spatial transcriptomics studies are even more expensive ($5,000–$15,000 per tissue section).

How sequencing data flows into clinical and research workflows

The raw output of any DNA sequencing method is a FASTQ file containing read sequences and quality scores. From there, the standard pipeline:

1. Alignment — reads are mapped to a reference genome (typically GRCh38 for human in 2026, GRCh37 still in clinical legacy systems) using aligners like BWA-MEM, minimap2 (long reads), or STAR (RNA-seq). Output: BAM/CRAM.
2. Variant calling — identify positions where the sample differs from reference. Standard tools: GATK HaplotypeCaller, DeepVariant, DRAGEN, Strelka2 (somatic). Output: VCF.
3. Annotation — variants are annotated with predicted functional consequences (Ensembl VEP, ANNOVAR) and matched to disease databases (ClinVar, OMIM, COSMIC).
4. Interpretation — clinical variants are classified against the ACMG/AMP guidelines for pathogenicity. Research variants are filtered for downstream analysis.
5. Federation + harmonization — for population-scale studies, harmonized variants are loaded into federated TRE substrates linked to clinical phenotypes via OMOP CDM, FHIR R4, or GA4GH VRS for cross-source comparability.

In production federal-health programs — the NIH All of Us Researcher Workbench, Genomics England’s GEL data environment, the NHLBI BioData Catalyst — sequencing data lives behind federated analytics layers. Researchers bring their pipelines to where the data lives instead of moving petabytes of FASTQ/BAM files around. Tools like Nextflow (Lifebit co-invented Nextflow, Di Tommaso et al. Nature Biotechnology 2017) make these pipelines portable across sites.

Frequently asked questions

What are the main DNA sequencing methods?
The four main DNA sequencing methods in 2026 are: (1) Sanger sequencing for single-gene targeted work and clinical confirmation; (2) short-read next-generation sequencing (NGS) — dominated by Illumina — for whole-genome, whole-exome, and panel sequencing at scale; (3) long-read sequencing — PacBio HiFi and Oxford Nanopore — for structural variants, complex regions, and methylation detection; and (4) single-cell and spatial sequencing for per-cell or anatomically-resolved measurements.

What is the difference between Sanger sequencing and NGS?
Sanger sequencing reads one DNA fragment at a time, producing high-accuracy reads of 800–1,000 base pairs. Next-generation sequencing (NGS) reads millions to billions of DNA fragments in parallel, with shorter reads (typically 150–300 bp) but vastly higher total throughput. Sanger is preferred for single-gene targeted work and clinical confirmation. NGS is preferred for whole-genome, whole-exome, and any application requiring high throughput.

How much does whole-genome sequencing cost in 2026?
A typical 30× whole-genome sequencing run on Illumina platforms costs $200–$600 at major sequencing centers (Broad, Sanger, BGI) in 2026, down from $1,000+ in 2014. Long-read WGS (PacBio HiFi or Oxford Nanopore Q20+) costs $700–$1,400 per genome at similar coverage. Downstream analysis (variant calling, annotation, interpretation) and storage often exceed the sequencing cost itself.

What is long-read sequencing used for?
Long-read sequencing produces reads of 10 kb to 4+ Mb — large enough to span structural variants, repetitive regions, and complex genomic features that short-read NGS cannot resolve. Used for: structural variant detection, haplotype phasing, repeat-expansion disorders (Huntington’s, fragile X), de novo genome assembly, direct methylation detection without bisulfite conversion, and real-time pathogen sequencing in clinical metagenomics.

What is the difference between PacBio and Oxford Nanopore?
PacBio HiFi uses circular consensus sequencing on the Revio/Sequel platforms to produce highly-accurate (99.9%) long reads of 15–25 kb. Oxford Nanopore (MinION, PromethION) drives single-stranded DNA through protein nanopores and reads the resulting current changes, producing reads up to 4+ Mb at lower per-base accuracy (95–99% depending on chemistry, ~99% at Q20+). PacBio is preferred when accuracy matters most. Nanopore is preferred for the longest reads, real-time analysis, and field-deployable sequencing.

Is NGS the same as whole-genome sequencing?
No — NGS is a family of technologies, while whole-genome sequencing (WGS) is one application of NGS. NGS encompasses any massively-parallel sequencing approach, including WGS, whole-exome sequencing (WES), targeted gene panels, RNA-seq, ChIP-seq, ATAC-seq, microbiome sequencing, and others. WGS specifically means sequencing the entire genome (typically at 30× coverage for clinical use).

What is single-cell DNA sequencing?
Single-cell DNA sequencing measures genomic variation within individual cells, rather than the average across millions of cells in a sample. Used to detect clonal evolution in cancer (which subpopulations are driving resistance?), somatic mutation profiling in development, and pre-implantation embryo testing. The 10x Genomics Chromium and Mission Bio Tapestri are the leading commercial platforms in 2026.

How is DNA sequencing data analyzed?
The standard pipeline: (1) align reads to a reference genome with BWA-MEM (short reads) or minimap2 (long reads); (2) call variants using GATK, DeepVariant, DRAGEN, or Strelka2; (3) annotate variants with Ensembl VEP and ANNOVAR; (4) match to clinical databases (ClinVar, OMIM, COSMIC) for interpretation. In 2026 these pipelines run as containerized Nextflow or Snakemake workflows, increasingly inside federated TRE environments where the analysis goes to the data rather than the data being centralized.

How Lifebit fits into the sequencing data pipeline

Lifebit’s federated trusted research environment is the platform layer for sequencing data in federal health programs and national genomics initiatives. The Data Factory Omics Module runs containerized GATK / BWA / DeepVariant / minimap2 pipelines orchestrated through Nextflow, harmonizes variants via GA4GH VRS, links to clinical phenotypes through OMOP CDM v5.4, and exposes federated analytics to researchers without moving FASTQ/BAM files between institutions.

The same platform is in production today at the NIH National Library of Medicine, Genomics England, the Danish National Genome Center, CanPath (Canadian national TRE), and the Cambridge Biomedical Research Centre — where it enabled the finding that 27% of breast cancer patients could be treated differently (Black D et al. Lancet Oncology 2025). Lifebit co-invented Nextflow, the workflow framework that has become the standard for reproducible bioinformatics pipelines at population scale.

If you’re scoping sequencing data infrastructure for a federal program — NIH, ARPA-H, FDA, VA, CMS — or a national-scale genomics initiative, book a 30-minute scoping call and we’ll walk through the architecture for your specific scale and data types.

Sources:
– Nextflow — Di Tommaso et al., Nature Biotechnology 2017
– Broad Institute Genomics Platform — sequencing services
– Illumina sequencing platforms
– PacBio HiFi sequencing
– Oxford Nanopore Technologies
– 10x Genomics Chromium platform
– GA4GH Variation Representation Specification (VRS)
– ACMG/AMP variant classification guidelines
– Cambridge Biomedical Research Centre — breast cancer cohort, Black D et al. Lancet Oncology 2025

Last updated: May 11, 2026