Single-Cell Full-Length Transcriptome Sequencing Service
CD Genomics provides a Single-Cell Full-Length Transcriptome Sequencing Service that combines 10x Genomics single-cell partitioning with long-read sequencing to capture complete transcript isoforms from individual cells. Unlike conventional 3'-end scRNA-seq — which sequences only the terminal fragment of each transcript and collapses isoforms into gene-level counts — full-length sequencing reads through entire cDNA molecules, preserving splice junctions, alternative start and end sites, and fusion transcript boundaries. The result is transcript-level resolution at single-cell granularity: isoform switching across cell types, alternative splicing events driving cell-state transitions, and cell-type-specific fusion transcripts that gene-level analysis cannot resolve.
- Long-read sequencing captures full-length transcript isoforms — not just 3' ends
- Isoform-level quantification, alternative splicing, and fusion transcript detection per cell
- Compatible with 10x Genomics single-cell partitioning followed by Nanopore or PacBio long-read sequencing
- Bridges the gap between single-cell resolution and full-length transcript structure
Why Full-Length Single-Cell Transcriptome Sequencing?
Standard 3' scRNA-seq platforms (10x Chromium, BD Rhapsody) count transcripts by capturing poly-A tails and sequencing short tags from the 3' end. This approach excels at gene-level quantification — identifying which genes are expressed in which cells — but systematically loses information about transcript structure: which exons are included or skipped, where transcription starts and ends, and whether two gene fragments are fused into a chimeric transcript.
Full-length single-cell transcriptome sequencing solves these limitations by reading across the entire cDNA molecule. The key capabilities that distinguish it from 3' scRNA-seq include:
- Isoform-level analysis. Gene-level counts average together multiple transcript isoforms, but isoform switching — where one cell type expresses isoform A and another expresses isoform B of the same gene — is invisible to 3' methods. Full-length sequencing resolves and quantifies each isoform independently.
- Alternative splicing detection. Exon inclusion/exclusion, intron retention, and alternative donor/acceptor site usage are directly captured from full-length reads, enabling cell-type-specific splicing analysis.
- Fusion transcript discovery. Chimeric transcripts formed by gene fusions are detectable because the long read spans the fusion breakpoint, unlike short-read tags that map ambiguously.
- Novel isoform and lncRNA identification. Full-length reads can reveal previously unannotated transcript isoforms, non-coding RNAs, and transcript structures that reference annotations miss.
- Allele-specific isoform expression. When reads span heterozygous variants, full-length sequencing can determine which allele produces which isoform — information relevant to imprinting, allelic imbalance, and cancer biology.
Long-Read Technology Overview
Our service supports two complementary long-read sequencing platforms:
Oxford Nanopore Technologies (ONT). Nanopore sequencing passes single-stranded cDNA through a protein nanopore embedded in a membrane, measuring changes in ionic current as each nucleotide passes through. Because read length is determined by the cDNA molecule, not by chemistry — the polymerase synthesizes until it reaches the end — typical read N50 values of 1–2 kb capture most full-length transcripts. ONT platforms (PromethION, GridION) generate sufficient throughput for thousands of single cells per run, with real-time data availability enabling early QC assessment.
PacBio HiFi Sequencing. PacBio's circular consensus sequencing (CCS) produces high-accuracy long reads (HiFi reads, >99.9% accuracy) by repeatedly sequencing a circularized cDNA template. While read lengths are shorter than ONT (typically 1–3 kb), the exceptional base-level accuracy makes HiFi reads well-suited for precise isoform boundary definition, SNP detection within transcripts, and de novo isoform discovery without a reference genome.
Platform selection guide. ONT is typically chosen for projects prioritizing maximum read length, novel isoform discovery, and fusion transcript detection where the fusion breakpoint may be far from the poly-A tail. PacBio HiFi is preferred when base-level accuracy is critical — for example, detecting single-nucleotide variants within specific isoforms, or distinguishing nearly identical isoforms that differ by a few nucleotides at a splice junction. Our team advises on platform selection during study design based on your biological question and required analytical resolution.
Full-Length scRNA-seq Workflow
CD Genomics manages the complete single-cell full-length transcriptome sequencing workflow.
- Sample preparation and single-cell partitioning
Fresh or cryopreserved single-cell suspensions are loaded onto a 10x Genomics Chromium controller, where individual cells are encapsulated in gel beads-in-emulsion (GEMs). Within each GEM, cells are lysed and mRNA is reverse-transcribed into barcoded cDNA — each cDNA molecule carries a cell-specific barcode and a transcript-specific molecular tag, enabling per-cell and per-molecule identification.
- Full-length cDNA amplification and QC
Barcoded cDNA is pooled and amplified. cDNA quality is assessed by fragment analyzer (Agilent Bioanalyzer or TapeStation) to confirm the expected size distribution (~1–3 kb) and sufficient yield. Samples with degraded or truncated cDNA profiles are flagged for review.
- Long-read library preparation and sequencing
Amplified cDNA is converted into a long-read sequencing library. For ONT, cDNA is end-repaired, adapter-ligated, and sequenced on PromethION or GridION flow cells. For PacBio, cDNA is circularized and processed for HiFi sequencing on a Sequel II or Revio system. Library QC includes concentration measurement and fragment size verification.
- Data processing and isoform reconstruction
Raw long reads are basecalled (ONT: Guppy/Dorado; PacBio: SMRT Link → CCS). Cell barcodes and molecular tags are extracted, demultiplexed to single-cell resolution, and aligned to the reference genome/transcriptome using minimap2 or pbmm2. Full-length isoforms are reconstructed and quantified using FLAIR, SQANTI3, or Bambu.
- Isoform analysis and biological interpretation
Alternative splicing events, isoform switching between cell types, fusion transcripts, and novel isoforms are identified. Results are interpreted in the biological context of the project and delivered with publication-ready figures, as detailed in the Bioinformatics Analysis section.
Full-Length Transcriptome Sample Requirements
Full-length single-cell transcriptome sequencing requires high-quality single-cell suspensions with high viability and minimal RNA degradation. Full-length cDNA synthesis is sensitive to RNA integrity — degraded RNA produces truncated cDNA molecules that underrepresent the 5' ends of transcripts.
| Sample Type | Requirement | Storage & Shipping | Notes |
|---|---|---|---|
| Fresh single-cell suspension | ≥ 100,000 viable cells; viability ≥ 85% | Process immediately; ≤ 2 hours at 4°C | Preferred input. Minimize time from dissociation to loading. |
| Cryopreserved single cells | ≥ 200,000 cells/vial; post-thaw viability ≥ 70% | −80°C or LN2; ship on dry ice | Thaw and assess viability before loading. |
| Fresh tissue (for nuclei isolation) | ≥ 30 mg; flash-frozen | Snap-freeze in LN2 immediately after collection; ship on dry ice | For tissues where intact cell dissociation is challenging; nuclei isolation preserves nuclear transcripts. |
| Pre-amplified full-length cDNA | ≥ 50 ng; fragment size 800–2,000 bp | −20°C; ship on dry ice | For customers who have already generated 10x single-cell cDNA. |
RNA integrity is critical for full-length transcript representation. Samples with extensive RNA degradation will produce 3'-biased cDNA that defeats the purpose of full-length sequencing. We recommend running an aliquot on a Bioanalyzer to assess RNA integrity before committing to full-length library preparation.
Bioinformatics Analysis
Full-length single-cell transcriptome data requires specialized bioinformatics pipelines distinct from 3' scRNA-seq analysis.
Standard analysis
- Long-read preprocessing: basecalling, cell barcode extraction, molecular tag demultiplexing, read filtering by quality and length
- Alignment and isoform reconstruction: reads aligned to reference genome/transcriptome; full-length isoforms identified and classified (known, novel, fusion)
- Isoform quantification: per-cell, per-isoform count matrices with transcript-level resolution
- Cell clustering and annotation: clustering based on isoform-level expression profiles; cell-type annotation using canonical markers
- Alternative splicing analysis: per-cell quantification of alternative splicing events (exon skipping, intron retention, alternative 5'/3' splice sites, mutually exclusive exons)
- Isoform switching analysis: identification of isoforms that are differentially used between cell types or conditions
- Fusion transcript detection: per-cell identification of chimeric transcripts and fusion gene candidates
- Differential isoform expression: statistical comparison of isoform usage between conditions or cell types
Advanced analysis (optional)
- Allele-specific isoform expression: isoform quantification stratified by parental allele using heterozygous SNPs within reads
- Novel isoform discovery and annotation: de novo transcript assembly and characterization of unannotated isoforms
- Long non-coding RNA (lncRNA) identification: isoform-level profiling of lncRNAs at single-cell resolution
- Multi-omics integration: correlation of isoform expression with matching scATAC-seq or spatial transcriptomics data
Full-Length Transcriptome Applications
Full-length single-cell transcriptome sequencing enables isoform-resolution discovery across multiple research areas.
Cancer biology and tumor heterogeneity
Detects oncogenic fusion transcripts at single-cell resolution, identifies tumor cell subclones by isoform signatures rather than gene-level expression alone, and reveals alternative splicing programs associated with drug resistance or metastasis.
Developmental biology
Isoform switching during cell differentiation — where progenitor cells and terminally differentiated cells express different isoforms of the same developmental regulators — is directly measured from full-length reads, capturing isoform-level regulatory programs that gene-level analysis overlooks.
Neuroscience
The nervous system exhibits extensive alternative splicing, with many neuronal genes producing dozens of isoforms. Full-length single-cell sequencing resolves neuron-subtype-specific isoform expression, splicing programs in synaptic genes, and isoform changes in neurodegeneration. See also: Spatial Omics Solutions for Neuroscience.
Immunology and immune repertoire
Full-length sequencing captures the complete variable regions of BCR and TCR transcripts at single-cell resolution, enabling paired heavy-light chain reconstruction, somatic hypermutation analysis, and clonotype tracking without separate targeted enrichment.
Plant biology
Plant genomes contain extensive gene families and polyploidy; full-length isoform sequencing at single-cell resolution resolves homeolog-specific isoform expression, tissue-specific splicing, and novel transcript discovery in non-model plant species.
Deliverables
- Processed sequencing data
- Basecalled and demultiplexed long reads per cell (FASTQ), aligned reads (BAM)
- Isoform expression matrix
- Per-cell, per-isoform count matrix with transcript annotations (.csv, .h5ad, .rds)
- QC report
- Per-cell QC metrics (read count, alignment rate), fragment size distribution, cell viability and cDNA quality assessment
- Full-length transcriptome analysis report
- Cell clustering visualization (UMAP), isoform-level feature plots, alternative splicing event summaries, isoform switching heatmaps, differential isoform expression results, and fusion transcript tables
- Publication-ready figures
- UMAP plots by cluster and isoform expression, sashimi plots, isoform switching volcano plots, and fusion transcript diagrams (300+ dpi TIFF/PNG; editable PDF/SVG)
- Reproducible analysis code
- Fully documented scripts with software versions and parameter settings
Case Study: High-Throughput Single-Cell Full-Length Isoform Sequencing Using PacBio CCS
Source: Shi ZX, Chen ZC, Zhong JY, et al., Nature Communications, 2023
Background
Single-cell full-length isoform sequencing holds the promise of revealing transcript diversity at cellular resolution — capturing which isoforms are expressed in each cell, how alternative splicing varies across cell types, and what novel transcripts exist beyond reference annotations. However, PacBio single-cell isoform sequencing has historically been limited by low throughput: artifact cDNAs from incomplete reverse transcription consume sequencing capacity, and short library preparation workflows produce a small number of full-length reads per SMRT Cell.
Methods
The authors developed HIT-scISOseq, a method that improves single-cell PacBio isoform sequencing throughput in two key ways. First, artifact cDNAs are selectively removed before library preparation, ensuring sequencing capacity is dedicated to genuine full-length transcripts. Second, multiple cDNAs are concatenated into longer molecules for PacBio circular consensus sequencing (CCS), dramatically increasing the number of full-length isoform reads per SMRT Cell without sacrificing the >99.9% base accuracy characteristic of HiFi reads. The method was validated on single-cell cDNA libraries from human PBMCs and mouse brain tissue.
Results
HIT-scISOseq achieved a >15-fold increase in single-cell isoform sequencing throughput compared to standard PacBio Iso-Seq, enabling isoform-level profiling of thousands of single cells per SMRT Cell. The method detected known and novel isoforms across cell types, accurately quantified isoform switching between immune cell subpopulations (e.g., CD14+ monocytes vs. CD4+ T cells), and identified cell-type-specific alternative splicing events. Isoform-level clustering resolved cell types that were not distinguishable by gene-level expression alone.
Conclusion
This study demonstrates that PacBio CCS-based single-cell full-length isoform sequencing can achieve the throughput necessary for large-scale single-cell experiments, while preserving the base-level accuracy required for precise isoform boundary definition and novel transcript discovery. The technical principles — artifact removal for cleaner libraries and cDNA concatenation for higher throughput — directly inform the sample preparation and sequencing strategies CD Genomics employs in its single-cell full-length transcriptome sequencing service.
Adapted from Shi et al., Nature Communications, 2023, doi:10.1038/s41467-023-38324-9.
Comparison: Full-Length vs. 3' Single-Cell RNA-Seq
Choose the appropriate approach based on your research question and required resolution.
| Dimension | Full-Length scRNA-seq | 3' scRNA-seq (10x, BD Rhapsody) |
|---|---|---|
| Read coverage | Entire cDNA molecule | 3' terminal fragment only |
| Resolution | Isoform-level | Gene-level |
| Alternative splicing | Directly detected per cell | Not accessible |
| Fusion transcripts | Detectable (read spans breakpoint) | Not reliably detectable |
| Novel isoforms | Discoverable | Not accessible |
| Sequencing platform | ONT or PacBio | Illumina short-read |
| Gene quantification | Yes (aggregated from isoforms) | Yes (default mode) |
| Input requirement | Higher RNA integrity needed | Tolerates moderate degradation |
| Cost per cell | Higher | Lower |
Choose full-length sequencing when your research question requires isoform-level resolution, splicing analysis, fusion detection, or novel transcript discovery. Choose 3' scRNA-seq when gene-level quantification, cell clustering, and marker-based annotation are sufficient.
Frequently Asked Questions (FAQ)
References
- Shi ZX, Chen ZC, Zhong JY, et al. "High-throughput and high-accuracy single-cell RNA isoform analysis using PacBio circular consensus sequencing." Nature Communications, vol. 14, 2023, 2631.
- Lu H, Giordano F, Ning Z. "Oxford Nanopore MinION sequencing and genome assembly." Genomics Proteomics Bioinformatics, vol. 14, no. 5, 2016, pp. 265–279.
- Wenger AM, Peluso P, Rowell WJ, et al. "Accurate circular consensus long-read sequencing improves variant detection and assembly of a human genome." Nature Biotechnology, vol. 37, 2019, pp. 1155–1162.
- Tian L, Jabbari JS, Thijssen R, et al. "Comprehensive characterization of single-cell full-length isoforms in human and mouse with long-read sequencing." Genome Biology, vol. 22, 2021, 310.
- Lebrigand K, Magnone V, Barbry P, Waldmann R. "High throughput error corrected Nanopore single cell transcriptome sequencing." Nature Communications, vol. 11, 2020, 4025.
- Volden R, Palmer T, Byrne A, et al. "Improving nanopore read accuracy with the R2C2 method enables the sequencing of highly multiplexed full-length single-cell cDNA." Proceedings of the National Academy of Sciences, vol. 115, no. 39, 2018, pp. 9726–9731.
