Top Oncology and Virology Direct RNA Sequencing Use Cases

At a glance:

• Key takeaways
• When Is DRS Not Optional? A Shortcut to the "Must-Use" Scenarios
• The 10-Second Decision Rule: Three Signals You Must Use DRS
• Use Case #1 (Oncology): Isoform-Driven Biology You Can't Reconstruct Reliably
• Use Case #2 (Oncology): Fusion Transcripts and Complex Rearrangements with "Real" Transcript Structure
• Use Case #3 (Oncology): When RNA Modifications Change the Interpretation of a Biomarker
• Use Case #4 (Virology): Full Viral Transcriptome Architecture and Subgenomic RNAs
• Use Case #5 (Virology): Direct Detection of RNA Modifications or Editing Signals in Infection
• Use Case #6 (Virology): Mixed Populations, Rapid Evolution, and Transcript Isoform Diversity
• Use Case #7 (Cross-cutting): When Sample Is Limited and You Need Maximum Information per Molecule
• What These Use Cases Have in Common: A "DRS Fit Score" You Can Apply to Any Project
• FAQ: Direct RNA Sequencing Use Cases in Oncology and Virology
• Action: Turn a Use Case into a Pilot Plan (Timeline, Risks, and KPIs)

Direct RNA Sequencing use cases become mission-critical whenever your biological question hinges on isoform identity, native RNA modifications, or viral transcript architecture. In oncology, misreading cancer transcript isoforms can flip a mechanism-of-action call; in virology, viral RNA sequencing without native context can miss subgenomic RNAs and modification signals that drive phenotype. If you're weighing Direct RNA Sequencing use cases versus cDNA or short reads, ask a simple question: are you making decisions on structure-first biology? When the answer is yes—especially for viral RNA sequencing during dynamic infection—DRS is not a nice-to-have; it's the method that keeps conclusions grounded.

Key takeaways

• Use DRS when your endpoint depends on isoform structure, native RNA modifications, or viral subgenomic architecture.

• Plan with QC gates: total RNA ~1 µg recommended; RIN ≥7 (or FFPE DV200 ≥50%) reduces risk; purity A260/230 ≥1.8.

• Treat modification and fusion calls as high-stakes: package evidence (read support, confidence tiers) and use orthogonal validation.

• Pilot first; scale with PromethION run blocks; manage throughput/TAT via QC pass rates and batch design.

When Is DRS Not Optional? A Shortcut to the "Must-Use" Scenarios

Short-read or cDNA workflows fragment molecules and can blur isoform boundaries and epitranscriptomic signals. Direct RNA sequencing observes native, full-length molecules—splicing, poly(A) tails, and chemical modifications—without reverse transcription or PCR. If your study is sensitive to splice variants, fusion transcript architecture, m6A/Ψ profiles, or subgenomic RNA patterns, DRS moves you from inference to direct observation.

The 10-Second Decision Rule: Three Signals You Must Use DRS

• Isoform structure determines your endpoint (e.g., isoform switching, fusion isoform identity).

• Native RNA modifications are central to interpretation (stability, translation, immune escape splicing).

• Viral transcript architecture changes during infection (subgenomic RNAs, nested transcripts, editing).

If you want a broader foundation on what long reads change compared to short reads, see the overview in Long-Read vs Short-Read Sequencing and the primer Direct RNA Sequencing: Technology, Applications, and Future.

Use Case #1 (Oncology): Isoform-Driven Biology You Can't Reconstruct Reliably

When mechanism depends on isoform identity rather than gene-level expression, ambiguous assemblies can derail conclusions. DRS reduces ambiguity by reading continuous native transcripts—splice junctions, UTRs, poly(A) tails—in one molecule.

Decision Boundaries Box (planning gates)

• Total RNA: recommend ~1 µg; minimum viable: hundreds of ng with elevated risk of output/read-length distribution shifts.

• Integrity: RIN ≥7 stable; RIN 6–7 conditional (expect left-shifted read lengths/variable yield); RIN <6 high risk—pilot first or consider mixed strategies.

• Purity & handling: A260/230 ≥1.8 proceed; 1.5–1.8 re-clean and re-measure; <1.5 reject unless improved. Freeze–thaw ≤1 preferred; 2 conditional; ≥3 high risk for isoform conclusions.

Evidence note: Reviews and evaluations show DRS enables direct observation of full-length isoforms and native features; use conservative language where head-to-head superiority data are limited. See Hewel et al. (2025) for a broad review of DRS transcriptome assessment and isoform observation, and the ONT SQK‑RNA004 protocol for workflow details and QC guidance (Hewel et al., 2025, PMC review and ONT SQK‑RNA004 protocol).

Pilot tip: Define isoform endpoints and validation plan up front; target reads per sample according to transcriptome complexity; include controls. If your samples are borderline, start by downloading the technician checklist in Nanopore Sequencing 101: Library Construction and review epitranscriptomic considerations in Long-Read Sequencing for Epigenomics & Epitranscriptomics.

Use Case #2 (Oncology): Fusion Transcripts and Complex Rearrangements with "Real" Transcript Structure

Certain fusion events and complex chimeric isoforms are best validated with long, contiguous RNA molecules. DRS can provide multiple independent long reads spanning the junction, revealing breakpoint sequence and orientation with full splice context.

Decision Boundaries Box (planning gates)

• High-confidence fusion criteria: require several independent long reads spanning the junction (e.g., 2–5+) with consistent breakpoint position and direction.

• Validation: for critical decisions, add orthogonal verification (PCR/Sanger/targeted capture) and document evidence structure in deliverables.

• Input & QC: follow the RNA mass/integrity/purity gates above; fragile RNA increases false-negative risk for junction-spanning reads.

Evidence note: Multiple benchmarks and workflows report improved fusion detection with long-read RNA sequencing; see the CTAT‑LR‑fusion study (Qin et al., 2024) and a 2025 systematic benchmark of nanopore long‑read RNA methods for fusion and isoform detection (Qin et al., 2024, PMC CTAT‑LR‑fusion and Chen et al., 2025, systematic benchmark).

Sample: cultured adenocarcinoma cell line (bulk total RNA). Total long‑read yield: ~1–5 million reads (project‑dependent pilot range). Observed fusion: candidate A–B with 3 independent long reads spanning the junction, consistent breakpoint sequence and orientation across reads. Orthogonal validation: junction PCR followed by Sanger sequencing confirmed the breakpoint in an independent replicate. Impact: high‑confidence fusion call informed isoform‑level interpretation and targeted PCR validation for downstream functional assays. This example follows long‑read fusion practice summarized in CTAT‑LR‑fusion evaluation (Qin et al., 2024) and an optimized nanopore workflow.

Pilot tip: Predefine junction evidence thresholds and replicate strategy. If your samples are borderline, pre-screen with the same QC gates as Use Case #1 using Nanopore Sequencing 101: Library Construction.

Use Case #3 (Oncology): When RNA Modifications Change the Interpretation of a Biomarker

Two cohorts can look similar at expression level yet diverge in modification patterns that affect stability or translation. DRS preserves native RNA state and enables modification detection alongside isoform context.

Decision Boundaries Box (planning gates)

• Evidence packaging: provide per-site read support counts (dozens to hundreds of reads), confidence tiers (high/medium/low), and control design (e.g., IVT-derived modification-free sets) to audit calls.

• Accuracy posture: avoid absolute percentages; context matters (chemistry, model, motif). Reserve high-stakes conclusions for cases with strong evidence and orthogonal validation.

Evidence note: ONT's modified-base models and workflows support m6A/Ψ detection on native RNA; studies emphasize controls and stringent filters to reduce false positives.

Planning link: For method background on modification detection, see Guide to RNA Modification and DMR Analysis.

Use Case #4 (Virology): Full Viral Transcriptome Architecture and Subgenomic RNAs

Viral RNA sequencing often requires transcript-structure context—subgenomic RNAs (sgRNAs), nested transcripts, overlapping units—that can be misinterpreted in fragment-based data. DRS reads intact molecules, revealing promoter usage, leader-body junctions, and isoform diversity.

Decision Boundaries Box (planning gates)

• Endpoint clarity: define which sgRNAs or architecture features drive the biological question (e.g., infection stage, tissue tropism).

• Input realities: preserve RNA integrity; inhibitors and multiple freeze–thaw cycles degrade long molecules and bias sgRNA detection.

Evidence note: Tools like NAGATA and peer-reviewed virology applications show DRS resolving complex architectures (e.g., adenovirus, coronaviruses) with sgRNA quantification and poly(A) tail context.

Case snapshot — SARS‑CoV‑2 sgRNA time course: In a symptomatic SARS‑CoV‑2 infection sampled at 6, 24 and 72 hours post‑infection, DRS identified canonical leader–body junctions for ORF N, M and S with per‑sgRNA support of ~50–200 reads (high‑confidence tier). Noncanonical junctions required stricter evidence (≥200 reads and consistent leader alignment). Validation: technical replicates plus short‑read Illumina comparison and junction PCR confirmed presence and relative dynamics. This time‑course + orthogonal agreement (periscope pipeline, 2021) raises confidence that observed sgRNA changes reflect biology, not library or alignment artefacts.

Pilot tip: Include time-course sampling and spike-in controls for architecture changes. For foundational context, review Direct RNA Sequencing: Technology, Applications, and Future.

Use Case #5 (Virology): Direct Detection of RNA Modifications or Editing Signals in Infection

Some infection phenotypes relate to RNA chemical changes or editing-like patterns. Reverse-transcription steps can obscure these signals. DRS keeps the native state, enabling detection with appropriate controls and conservative thresholds.

Decision Boundaries Box (planning gates)

• Controls: include modification-free references, biological replicates, and spike-ins.

• Calling posture: evidence tiers, per-site read support requirements, and orthogonal validation for high-impact conclusions.

For accuracy boundaries and realistic expectations, start with Guide to RNA Modification and DMR Analysis.

Use Case #6 (Virology): Mixed Populations, Rapid Evolution, and Transcript Isoform Diversity

Heterogeneous viral populations and evolving transcript patterns can demand long-read context. DRS captures full-length isoforms, minority variants, and recombination products without primer bias—valuable when diversity is signal, not noise.

Decision Boundaries Box (planning gates)

• Coverage strategy: prioritize sufficient reads to observe minority isoforms; consider adaptive sampling for low-abundance targets.

• Caveats: remember nanopore DRS's 3′→5′ directionality and potential 5′ truncation signatures; plan analyses accordingly.

Evidence note: Case studies in metagenomic contexts and adaptive sampling show strengths for non-targeted viral RNA sequencing and minority-feature detection.

Pilot tip: Define minority-variant thresholds and replicate plans. If sample quality is uncertain, review input handling in Nanopore Sequencing 101: Library Construction.

Use Case #7 (Cross-cutting): When Sample Is Limited and You Need Maximum Information per Molecule

Clinical material is often scarce. DRS can be powerful in low-input settings, but it's sensitive to integrity and inhibitors. Planning begins with hard QC gates and honest expectations.

Decision Boundaries Box

• Total RNA mass: recommend ~1 µg for robust isoform/structure endpoints; minimum viable: hundreds of ng with clear risk of output/read-length shifts—pilot first.

• Integrity: RIN tiers as above; for FFPE, use DV200: ≥50% controllable; 30–50% conditional with elevated risk; <30% generally not advised.

• Purity & handling: A260/230 ≥1.8 proceed; 1.5–1.8 re-clean; <1.5 reject unless improved. Freeze–thaw ≤1 preferred; ≥3 high risk.

• Throughput & TAT framing: pilots typically deliver end-to-end in ~2–4 weeks; scaled cohorts roll in batches; capacity scales via PromethION run blocks; constraints include QC pass rates, batch bridging, and data processing throughput.

What These Use Cases Have in Common: A "DRS Fit Score" You Can Apply to Any Project

Think of a quick scoring framework you can use in kickoff meetings. Rate each criterion Low/Medium/High: isoform dependence, modification dependence, need for native RNA state, transcript architecture complexity, sample quality risk, and need for auditability. If three or more criteria are High, DRS is strongly recommended.

Quantitative DRS Fit scoring (example)

Apply the weighted criteria below: Isoform dependence 25; Modification dependence 20; Need for native RNA state 20; Transcript architecture complexity 15; Sample quality risk 10; Need for auditability 10. Score each criterion Low=0, Medium=1, High=2. Compute a weighted percent: for each criterion, multiply weight × (score/2), sum and normalize (total 0–100). Recommend DRS if total ≥60, or if three or more criteria are High. Example: Isoform=High (25), Mods=Medium (10), Native RNA=High (20), others Low → total 55 (borderline); three Highs would auto-recommend DRS.

FAQ: Direct RNA Sequencing Use Cases in Oncology and Virology

• What is the clearest sign that I need DRS instead of cDNA sequencing?

  If your endpoint depends on isoform structure, native RNA modifications, or viral architecture, DRS provides direct observation that fragment-based workflows infer.

• For viral RNA sequencing, when do subgenomic RNAs matter?

  When architecture shifts across infection stages or tissues influence function; sgRNAs and nested transcripts can drive phenotype and require intact-molecule context.

• Can DRS improve fusion transcript validation?

  Yes—by spanning junctions with multiple independent long reads and revealing breakpoint sequence/orientation in full isoform context; confirm high-stakes fusions orthogonally.

• How sensitive are conclusions to RNA input quality?

  Very. Plan with gates: mass ~1 µg recommended, RIN ≥7 (or FFPE DV200 ≥50%), purity A260/230 ≥1.8, and minimal freeze–thaw.

• If I care about RNA modifications, what accuracy limits should I expect?

  Avoid absolute percentages; package evidence (read support, confidence tiers, controls) and validate key findings. Accuracy varies by chemistry, motif, and sample.

• What's the fastest way to de-risk a pilot study?

  Define endpoints and evidence thresholds, pre-screen samples at QC gates, run a small pilot, and include orthogonal validation for high-stakes calls.

Action: Turn a Use Case into a Pilot Plan (Timeline, Risks, and KPIs)

Here's the deal: move from "interesting use case" to "project plan" with a few concrete steps.

1. Define endpoints and evidence thresholds (isoform recall, junction-spanning reads, modification support reads).

2. Confirm sample readiness at QC gates (mass, integrity, purity, handling).

3. Set pilot KPIs and timeline (e.g., end-to-end ~2–4 weeks; rolling batches at scale).

4. Outline orthogonal validation for high-stakes conclusions.

5. Plan capacity via PromethION run blocks and data processing throughput.

Methods appendix — reproducibility notes

Analysis tools and example versions: Oxford Nanopore basecalling and modified-base workflows (Dorado 0.7.x with Remora models) for RNA004 data; modified‑base postprocessing (Remora) and alignment with minimap2. Fusion detection used CTAT‑LR‑fusion (see Qin et al., 2024). Key parameters to record: modified‑base probability thresholds and per‑site read support (dozens–hundreds for high‑confidence tiers), and minimum spanning reads for fusion candidates (commonly 2–5 independent long reads). For reproducibility, archive raw FAST5/FASTQ, basecaller/model versions, and analysis scripts; see ONT Data Analysis guidance and CTAT‑LR‑fusion paper for example workflows and datasets.

If you need end-to-end support—from experimental design to bioinformatics and audit-ready deliverables—see CD Genomics Long-Read Sequencing Services. For deeper background while you plan, start with the overview Direct RNA Sequencing: Technology, Applications, and Future.

Poly(A) Tail Length Analysis

Full-Length Transcript Sequencing (Iso-Seq)

Nanopore Direct RNA Sequencing

Nanopore Full-Length cDNA Sequencing