Transcriptomics with Long-Read Sequencing — Full-Length Isoform Discovery & RNA Modification Analysis
Long-read sequencing captures complete transcript isoforms in single reads — no fragmentation, no assembly gaps. CD Genomics delivers the full spectrum of long-read transcriptomics: PacBio Iso-Seq with Kinnex throughput, ONT direct RNA with native modification detection, single-cell full-length profiling, and specialized RNA analysis — all on dual platforms with custom bioinformatics support.
Why researchers choose our transcriptomics services
- Full-length isoforms without assembly — PacBio HiFi ≥Q30
- Native RNA modification detection — ONT direct RNA sequencing
- Single-cell to spatial-temporal — complete transcriptomics coverage
- Custom bioinformatics with peer-reviewed tools
- Dual-platform (PacBio + ONT) — no vendor lock-in
At a glance:
- Why Long-Read Sequencing Transforms Transcriptomics
- PacBio vs ONT: Choosing the Right Platform
- Full-Length Isoform Sequencing Services
- Nanopore Direct RNA & cDNA Sequencing Services
- Specialized & Single-Cell Transcriptomics Services
- Bioinformatics & Data Analysis Pipeline
- Sample Requirements
- Case Study: Isoform Detection Benchmark
- Demo Results
- Frequently Asked Questions
Why Long-Read Sequencing Transforms Transcriptomics
Short-read RNA-seq has been the workhorse of transcriptomics for over a decade, but its fundamental limitation has never changed: RNA must be fragmented to 100–300 bp fragments, and transcript structures must be computationally reconstructed from these fragments. This assembly step introduces uncertainty — alternative splicing variants are often misassembled, fusion transcripts are missed entirely, and isoform-level quantification becomes unreliable when multiple isoforms share exonic regions.
Long-read sequencing eliminates this problem at its root. By reading entire RNA molecules in single contiguous reads — routinely 1–10 kb for PacBio HiFi and up to 100 kb for ONT — long-read platforms capture complete isoform structures without assembly. This means researchers can:
- Resolve full-length isoforms unambiguously. Each read represents a complete transcript, from 5′UTR through coding sequence to 3′UTR, revealing the exact exon combination without computational inference.
- Detect fusion transcripts with confidence. When a single read spans a fusion junction with flanking sequences from both partner genes, the fusion event is directly observed rather than statistically inferred.
- Discover novel isoforms and genes. Many transcripts — particularly those from unannotated loci, repetitive regions, or complex splice patterns — are invisible to short-read methods. Long-read sequencing routinely uncovers thousands of previously unknown isoforms per experiment.
- Capture allele-specific expression at isoform resolution. Phased long reads can distinguish maternal and paternal transcript isoforms, enabling direct measurement of allele-specific isoform usage.
For projects where transcript structure itself is the biological question — alternative splicing in disease, isoform switching during development, fusion transcript discovery in cancer — long-read sequencing is not just an upgrade; it is the only method that provides unambiguous answers. A 2024 benchmark study in Nature Communications confirmed that guided isoform detection methods achieve >90% precision and sensitivity — performance levels that short-read assembly cannot approach for full-length transcripts.
PacBio vs ONT: Choosing the Right Platform for Your Transcriptomics Project
The choice between PacBio and Oxford Nanopore for transcriptomics is not about which platform is "better" — it is about which platform matches your specific research question. Each technology has distinct strengths that make it the right answer for different experimental goals.
Platform Comparison for Transcriptomics
| Dimension | PacBio HiFi (Iso-Seq / Kinnex) | ONT Direct RNA | ONT cDNA |
| Read Length | 1–10 kb (HiFi); up to 20 kb with Kinnex | 1–5 kb typical; >10 kb possible | 1–10 kb typical; >100 kb possible |
| Base Accuracy | >99.9% (Q30+) — gold standard | 90–97% (Q10–Q15) | 90–98% (Q10–Q17) |
| RNA Modification Detection | Not preserved (cDNA-based) | Native detection through current signal perturbation | Not preserved (cDNA-based) |
| Throughput per Flow Cell | 15–30 Gb (Sequel IIe); 90 Gb (Revio) | 10–30 Gb (PromethION) | 10–50 Gb (PromethION) |
| Library Prep Complexity | SMRTbell ligation + cDNA synthesis; Kinnex concatenates 5 transcripts per SMRTbell | Direct RNA ligation — no reverse transcription, no PCR | cDNA synthesis + ligation; simpler than Iso-Seq |
| Input Requirement | 300 ng–2 µg total RNA | 500 ng polyA+ RNA recommended | 50–500 ng total RNA |
| Best For | Isoform discovery, annotation, novel isoform cataloging, fusion detection at highest accuracy | RNA modification mapping (m6A, pseudouridine, inosine), epitranscriptomics | Longest reads for complex loci, rapid transcriptome survey, low-input samples |
| Limitations | No modification information; reverse transcription may introduce artifacts | Lower accuracy; polyA+ enrichment required | No modification information; reverse transcription artifacts |
How to choose: Choose PacBio HiFi Iso-Seq when your priority is the most accurate full-length isoform catalog. Choose ONT Direct RNA when RNA modifications matter — m6A dynamics, pseudouridine sites, or any epitranscriptomic question. Choose ONT cDNA when you need the longest possible reads for complex loci or have limited input material. Many research programs benefit from using both platforms, and CD Genomics supports integrated dual-platform projects with unified bioinformatics reporting.
Full-Length Isoform Sequencing Services
Our PacBio-based isoform sequencing services provide the highest-accuracy full-length transcript data for unambiguous isoform discovery, annotation, and quantification.
Iso-Seq (Full-Length Transcript Sequencing)
Resolve complete transcript isoforms with PacBio HiFi accuracy — from 5′ end to 3′ polyA tail. Ideal for reference transcriptome construction, novel isoform discovery, and alternative splicing analysis across any eukaryotic species.
View Iso-Seq Service →
Kinnex Full-Length RNA-Seq
The Kinnex full-length RNA kit concatenates up to 5 full-length transcripts per SMRTbell, delivering 5–8× higher throughput than standard Iso-Seq at the same HiFi accuracy. Recommended for large-scale isoform projects and population-scale transcriptomics.
View Kinnex RNA-Seq →
Full-Length Transcriptome Profiling
The most comprehensive transcriptome view: Illumina short-read for quantitative gene expression plus PacBio or ONT long-read for full-length isoform resolution. Get both accurate quantification and complete isoform structures in a single project.
View Combined Profiling →
Spatial-Temporal Transcriptome Profiling
Integrate full-length isoform information with spatial and temporal dimensions. Capture how isoform usage changes across tissue regions and developmental time points — revealing splicing programs that bulk and short-read methods cannot resolve.
View Spatial-Temporal Service →
Nanopore Direct RNA & cDNA Sequencing Services
ONT-based transcriptomics services provide unique capabilities — native RNA modification detection and ultra-long reads — that complement our PacBio offerings.
Nanopore Direct RNA Sequencing
The only sequencing method that reads native RNA molecules without reverse transcription or PCR. Preserve and detect RNA modifications (m6A, pseudouridine, inosine, 5mC) directly from nanopore current signals. Essential for epitranscriptomics and modification-aware isoform analysis.
View Direct RNA Sequencing →
Nanopore Full-Length cDNA Sequencing
ONT-based cDNA sequencing delivers the longest achievable reads — routinely exceeding 10 kb and reaching 100 kb for complex loci. PCR-free library preparation provides full-length coverage of large genes and fusion transcripts that challenge other platforms.
View Full-Length cDNA →
Nanopore Direct lncRNA Sequencing
Long non-coding RNAs present unique challenges: often long (>5 kb), lowly expressed, and lacking polyA tails. Our direct lncRNA sequencing service adapts ONT direct RNA technology for lncRNA-specific workflows, capturing native lncRNA molecules with modifications intact.
View Direct lncRNA Sequencing →
Specialized & Single-Cell Transcriptomics Services
Beyond standard full-length transcriptomics, our specialized services address unique RNA types and single-cell resolution requirements.
Single-Cell Full-Length Transcriptome
Combine single-cell resolution with full-length isoform information using MAS-Seq or 10x Genomics workflows coupled with PacBio or ONT long-read sequencing. Reveals isoform-level heterogeneity, cell-type-specific splicing, and fusion transcripts invisible to short-read scRNA-seq.
View Single-Cell Service →
circRNA Full-Length Identification
Circular RNAs are covalently closed RNA circles with key roles in gene regulation and miRNA sponging. RNase R enrichment followed by long-read sequencing identifies full-length circRNA structures with precise back-splice junction mapping.
View circRNA Service →
Poly(A) Tail Length Analysis
Poly(A) tail length directly influences mRNA stability, translation efficiency, and localization. TAIL Iso-Seq and ONT-based approaches measure poly(A) tail length distributions at transcript-level resolution across conditions and developmental stages.
View Poly(A) Tail Analysis →
Nano tRNA Sequencing
tRNAs carry over 150 known chemical modifications invisible to standard RNA-seq. ONT direct RNA technology sequences full-length, native tRNAs — capturing modification patterns, isoacceptor usage, and quantitative abundance in a single workflow.
View tRNA Sequencing →
Nanopore Ribosome tRNA Sequencing
Ribosome-associated tRNAs represent the functionally active tRNA pool distinct from total cellular tRNA. Enriches ribosome-bound tRNAs and sequences them via ONT nanopore technology, revealing the translationally engaged tRNA repertoire under specific biological conditions.
View Ribosome tRNA Service →
Bioinformatics & Data Analysis Pipeline
The value of long-read transcriptomics depends as much on analysis as on sequencing. CD Genomics provides a standardized yet customizable bioinformatics pipeline built on independently benchmarked tools — the same tools that a 2024 Nature Communications study identified as best-in-class for long-read isoform detection.
Standard Analysis Pipeline
- Basecalling & QC: Raw signal to base calls (Dorado for ONT; SMRT Link for PacBio). Reads filtered by Q-score and length thresholds, with quality metrics reported per sample.
- Alignment: Splice-aware alignment to reference genome using minimap2. Mapping rate, read distribution, and coverage statistics reported.
- Isoform Detection & Classification: isoform quantification tool (guided mode) — top-performing tool in the benchmark — identifies full-length isoforms. SQANTI3 classifies as FSM, ISM, NIC, or NNC.
- Quantification: Bambu for context-aware transcript-level abundance (TPM + count matrices).
- Differential Analysis: DEXSeq/DRIMSeq for differential transcript usage and isoform expression.
- Functional Annotation: GENCODE/Ensembl/RefSeq annotation, ORF prediction, InterProScan/Pfam domain analysis.
- Reporting: Interactive HTML report plus raw data (BAM, GTF/GFF3, expression matrices, differential tables).
Custom analysis includes allele-specific isoform analysis, modification calling (ONT direct RNA), short-read integration, and comparative genomics. All deliverables include aligned BAM files, isoform annotations, expression matrices, differential analysis tables, and an interactive report.
Sample Requirements for Long-Read Transcriptomics
| Sample Type | Recommended Quantity | Minimum | Quality Requirement | Shipping |
| Total RNA (PacBio Iso-Seq) | 2 µg | 500 ng | RIN ≥ 7.0; 260/280 = 1.8–2.1 | Dry ice |
| Total RNA (ONT cDNA) | 1 µg | 50 ng | RIN ≥ 7.0; 260/280 = 1.8–2.1 | Dry ice |
| PolyA+ RNA (ONT Direct RNA) | 500 ng | 200 ng | RIN ≥ 7.0; no DNA contamination | Dry ice |
| Single-cell suspension | 10,000 cells | 1,000 cells | Viability > 85% | Fresh (4°C) or cryopreserved |
| Tissue (for RNA extraction) | 50–100 mg | 20 mg | Flash-frozen; –80°C storage | Dry ice |
| RNAlater-preserved | Per protocol | Per protocol | ≥5× volume RNAlater | Ambient (4°C preferred) |
Important: RNA integrity (RIN) is the most critical quality parameter. Degraded RNA produces truncated reads that compromise full-length isoform detection. We perform incoming QC on all samples before library preparation. For single-cell projects, cell viability >85% is essential — dead cells release RNases that compromise the entire experiment.
Case Study: Benchmarking Isoform Detection Tools for Long-Read Transcriptomics
Su Y, Yu Z, Jin S, et al. Comprehensive assessment of mRNA isoform detection methods for long-read sequencing data. Nature Communications. 2024;15:3972. (CC BY 4.0)
1. Background
As long-read RNA sequencing adoption accelerates, researchers face a practical challenge: which isoform detection tool should they use? Over a dozen computational methods exist for identifying mRNA isoforms from PacBio and ONT data, but systematic comparisons have been limited. The research team from Zhejiang University addressed this gap with the most comprehensive benchmark study to date, evaluating 13 tools across both major long-read platforms.
2. Methods
The team generated simulated long-read datasets with known ground-truth isoforms, spanning PacBio CLR, PacBio HiFi (CCS), and ONT R103 chemistries. They varied five key parameters — sequencing depth, transcriptome complexity, read length completeness, sequencing error rate, and reference annotation completeness — to assess tool performance under realistic conditions. This was supplemented with RNA-sequins (synthetic spike-in standards) and real experimental data from human embryonic stem cell (hESC) differentiation. Thirteen tools were evaluated in both guided and unguided modes.
Figure 2 from Su et al. 2024, Nature Communications (CC BY 4.0). Benchmark performance of 13 isoform detection tools across PacBio and ONT platforms, showing precision and sensitivity under varied experimental conditions.
3. Results
Isoform quantification tool (guided mode) achieved the highest overall accuracy, with >90% precision and sensitivity across platforms and conditions. Bambu (guided) closely followed, with particular strength in isoform-level quantification. StringTie2 was the fastest tool — completing analyses 5–50× faster than competitors. Critically, guided methods consistently outperformed unguided approaches, reducing false discovery rates from 30–60% to <10% in complex transcriptomes.
In the biological application — hESC transition from naïve to primed pluripotency — guided isoform quantification tool and Bambu identified a novel isoform of the RPL39L gene (RPL39L-L) with a previously unannotated 5′ exon, independently validated by RT-qPCR.
4. Conclusions
Carefully selected isoform detection tools — specifically isoform quantification tool and Bambu in guided mode — achieve >90% accuracy for full-length isoform identification from both PacBio and ONT data. At CD Genomics, we incorporate these benchmark-validated tools into our standard bioinformatics pipeline.
Demo Results: Long-Read Transcriptomics Data Showcase
The demo visualization below illustrates typical isoform-level analysis outputs from our long-read transcriptomics pipeline, including sashimi plots showing junction reads across multiple samples, SQANTI3 classification distributions (FSM 60–75%, ISM 15–25%, NIC 3–8%, NNC 2–5%), and read length histograms confirming full-length coverage with median read lengths of 1.5–3 kb.
Representative demo data: isoform sashimi plot, SQANTI3 classification distribution, and read length histogram from our long-read transcriptomics pipeline.
Frequently Asked Questions
Q: What is the difference between PacBio Iso-Seq and ONT direct RNA sequencing?
PacBio Iso-Seq sequences full-length cDNA with HiFi accuracy (>Q30), ideal for isoform discovery and annotation. ONT direct RNA sequencing reads native RNA molecules without reverse transcription, enabling direct detection of RNA modifications (m6A, pseudouridine, inosine) through nanopore current signals — but with lower per-base accuracy (Q10–Q15). Choose Iso-Seq for publication-grade isoform catalogs; choose ONT direct RNA when native modification information matters.
Q: How does long-read transcriptomics compare to Illumina short-read RNA-seq?
Short-read RNA-seq fragments RNA to 100–300 bp and computationally reconstructs transcripts, which often fails to resolve full-length isoforms, fusion transcripts, and complex splicing. Long-read sequencing reads entire transcripts in single contiguous reads — up to 10 kb or more — providing unambiguous isoform structures without assembly. Short-read offers higher throughput and lower cost for gene-level quantification; many projects combine both for comprehensive analysis.
Q: Can long-read sequencing detect RNA modifications?
Yes, but the capability is platform-dependent. ONT direct RNA sequencing detects modifications natively through characteristic nanopore current signal perturbations. PacBio Iso-Seq uses cDNA as input, so RNA modifications are not preserved. For modification-aware transcriptomics, ONT direct RNA is recommended. CD Genomics offers both platforms, enabling PacBio for definitive isoform annotation and ONT for modification mapping.
Q: What bioinformatics tools do you use for isoform detection?
We employ tools independently validated by the 2024 Nature Communications benchmark study. Our primary pipeline uses isoform quantification tool (guided mode) for isoform detection, with Bambu for quantification and StringTie2 for rapid processing. SQANTI3 classifies isoforms by structural category. This tool selection is evidence-based and continuously updated.
Q: What is the minimum RNA input for long-read transcriptomics?
For PacBio Iso-Seq, we recommend 2 µg total RNA (minimum 500 ng). For ONT direct RNA sequencing, 500 ng polyA+ RNA (minimum 200 ng). For ONT cDNA, as little as 50 ng total RNA can be sufficient. Single-cell projects require approximately 10,000 viable cells (>85% viability). Specific requirements depend on sample type — consult our team for a feasibility assessment.
Q: How long does a typical long-read transcriptomics project take?
Typical timelines: library preparation (3–5 business days), sequencing (2–7 days), data analysis (5–10 business days). Standard projects deliver analyzed data within 3–4 weeks from sample receipt. Expedited processing is available for time-sensitive projects.
Q: Can you analyze both known and novel isoforms?
Yes. Our pipeline identifies both annotated isoforms (matching reference databases) and novel isoforms (previously unannotated splice variants, fusion transcripts, or novel genes). Novel isoforms are classified using SQANTI3 into ISM, NIC, and NNC categories with quality metrics for prioritization. All detected isoforms are included in the final annotation and expression matrices.
Q: Do you offer combined short-read + long-read transcriptomics?
Yes. Our Full-Length Transcriptome Profiling service integrates Illumina short-read RNA-seq (for high-depth quantification) with PacBio or ONT long-read sequencing (for complete isoform resolution). This combined approach provides the most comprehensive transcriptome characterization — quantification depth plus structural completeness.
References
- Su Y, Yu Z, Jin S, et al. Comprehensive assessment of mRNA isoform detection methods for long-read sequencing data. Nature Communications. 2024;15:3972.
- Amarasinghe SL, Su S, Dong X, et al. Opportunities and challenges in long-read sequencing data analysis. Genome Biology. 2020;21(1):30.
- Monzó C, Liu T, Conesa A. Transcriptomics in the era of long-read sequencing. Nature Reviews Genetics. 2025;26:681–701.
For Research Use Only. Not for use in diagnostic procedures.