Single-cell sorting and sequencing experiments isolate individual cells from complex biological samples and resolve their genomic, transcriptomic, and epigenomic profiles at single-cell resolution. This technical article describes the complete experimental workflow—from sample preparation and cell sorting to library construction, sequencing, and bioinformatics analysis—along with quality control requirements, applicable sample types, and typical application scenarios in research and clinical laboratories.
Technical Overview
Single-cell sorting and sequencing refers to the integrated experimental process of physically isolating individual cells from a heterogeneous population and then generating molecular profiles—such as transcriptomes, genomes, epigenomes, or immune repertoires—from each isolated cell. Conventional bulk sequencing averages signals across thousands to millions of cells and masks rare subpopulations, whereas single-cell approaches reveal cell-to-cell heterogeneity, rare cell states, and clonal architecture that are invisible to bulk methods.
The workflow typically combines two core technologies. First, a cell sorting platform isolates viable single cells with high purity. Common sorting strategies include fluorescence-activated cell sorting (FACS), which uses fluorophore-conjugated antibodies and droplet-based electrostatic deflection to deposit individual cells into plates; magnetic-activated cell sorting (MACS), which enriches target populations via antibody-linked magnetic beads; and microfluidic droplet or nanowell platforms, which encapsulate single cells together with barcoded beads in oil-in-water emulsions. Second, a sequencing library construction chemistry captures messenger RNA, genomic DNA, or chromatin accessibility from each cell and labels molecules with cell-specific barcodes and unique molecular identifiers (UMIs), so that all reads can later be assigned back to their cell of origin.
Downstream sequencing is generally performed on short-read high-throughput platforms, followed by computational pipelines that perform alignment, barcode demultiplexing, UMI collapsing, quality filtering, dimensionality reduction, clustering, cell-type annotation, and differential expression or trajectory analysis. Laboratories supporting these experiments are encouraged to align their practices with internationally recognized frameworks, including ISO/IEC 17025 for general testing laboratory competence, ISO 20387 for biobanking of biological specimens, and the MIFlowCyt minimum information standard for flow cytometry data reporting.
Testing Samples
Primary Tissue Samples
Fresh solid tissues such as tumor biopsies, surgical resections, and animal organs are the most information-rich but most demanding sample type. Tissues must typically be transported under cold chain conditions in specialized preservation medium and enzymatically dissociated into single-cell suspensions within 24 to 48 hours of collection. The dissociation protocol—combining mechanical disruption with collagenase, dispase, or papain digestion—is optimized per tissue type to maximize yield while preserving surface epitopes and RNA integrity. Over-digestion or delayed processing rapidly reduces viability below the recommended threshold of 80 to 90 percent and introduces stress-response gene artifacts.
Blood and Immune Cells
Peripheral blood, bone marrow, cord blood, and isolated peripheral blood mononuclear cells (PBMCs) are well suited to single-cell experiments because they arrive naturally in suspension. Anticoagulated blood should be processed within a few hours of collection; alternatively, PBMCs can be cryopreserved in serum-containing medium with dimethyl sulfoxide and later thawed with controlled-rate procedures. Frozen PBMCs are acceptable provided post-thaw viability exceeds roughly 80 percent, which makes retrospective and multi-center immune profiling studies feasible.
Cultured Cells and Organoids
Cell lines, induced pluripotent stem cells, 3D spheroids, and patient-derived organoids are routinely submitted for single-cell characterization. Cultured cells require gentle dissociation using enzymes such as Accutase-type formulations or non-enzymatic buffers that maintain membrane integrity. Organoids must first be mechanically and enzymatically dispersed into single cells, with digestion conditions tuned to avoid clumping and doublets that would otherwise register as artificial "multicellular" barcodes.
Rare and Circulating Cells
Circulating tumor cells, fetal nucleated red blood cells, rare stem or progenitor populations, and antigen-specific lymphocytes represent extremely low-abundance targets. These samples benefit from enrichment strategies prior to droplet-based capture, including FACS index sorting into plates, magnetic pre-enrichment, or micromanipulation, ensuring that scarce biological material is not diluted below the detection limit of the platform.
Plant and Microbial Cells
Plant protoplasts prepared by cell-wall-degrading enzymes can be profiled to study differentiation, stress responses, and cell-type-specific regulatory programs. For microorganisms, single-cell approaches adapted to bacteria and fungi, such as combinatorial barcoding and split-pool methods, enable uncultured community profiling and population heterogeneity studies that culture-based methods cannot resolve.
Testing Items
A single-cell experiment can be configured to interrogate different molecular layers, and multiple modalities are frequently combined into multimodal panels:
- Single-cell RNA sequencing (scRNA-seq): quantifies whole-transcriptome or targeted gene expression per cell, supporting cell typing, state annotation, and differential expression analysis.
- Single-cell DNA sequencing (scDNA-seq): detects copy-number variations, single-nucleotide variants, and clonal structure in individual cells, particularly for tumor heterogeneity studies.
- Single-cell ATAC sequencing (scATAC-seq): maps open chromatin regions to infer regulatory elements and transcription factor activity at single-cell resolution.
- Single-cell immune repertoire profiling: characterizes paired T-cell receptor (TCR) and B-cell receptor (BCR) sequences together with gene expression to link clonotype identity with cell phenotype.
- Single-cell surface protein quantification: uses oligonucleotide-tagged antibodies to measure membrane protein abundance simultaneously with transcriptomes in the same cells.
- Cell quality assays: viability staining, cell counting, red blood cell and dead-cell depletion checks, and doublet rate estimation prior to library construction.
- Library and sequencing QC: library concentration and fragment-size distribution, sequencing depth, read saturation, barcode-rank knee plot evaluation, and demultiplexing statistics.
Testing Methods
The standard experimental workflow proceeds through six controlled stages, each with defined acceptance criteria.
Sample Receipt and Dissociation
Specimens are logged, inspected, and processed in a biosafety cabinet. Solid tissues undergo protocol-specific enzymatic and mechanical dissociation; suspension samples undergo red blood cell lysis, dead-cell removal, or depletion of unwanted lineages as required. Nuclease-free reagents and ribonuclease-free handling practices are mandatory throughout.
Cell Suspension Quality Control
Cell count, diameter distribution, aggregation, and viability are measured using automated cell counters with fluorescent viability dyes such as acridine orange/propidium iodide. Acceptance thresholds typically require viability above 80 to 85 percent, minimal clumping, and a cell concentration within the loading range specified for the chosen capture platform. Cell size must fall within the platform-compatible range to avoid clogging or capture failure.
Cell Sorting or Capture
Depending on the study design, cells are either sorted by FACS into 96- or 384-well plates containing lysis buffer—ideally with index sorting so that phenotype data accompany each well—or loaded onto a droplet-based microfluidic instrument for massively parallel capture of thousands to tens of thousands of cells per channel. FACS performance is verified daily using calibration beads, and sorting purity is checked by re-analysis of a sorted fraction. FACS documentation should follow the MIFlowCyt reporting standard to ensure reproducibility.
Library Construction
Captured cells are lysed, and mRNA is reverse-transcribed with cell barcodes and UMIs, or genomic DNA is amplified and fragmented with tagmentation-based chemistry, depending on the modality. After pooling and cleanup, libraries are assessed for concentration using fluorometric assays and for fragment size using capillary electrophoresis. Combinatorial dual indexes are applied to prevent index misassignment across pooled runs.
Sequencing
Libraries are sequenced on a short-read high-throughput instrument with read configurations appropriate to the assay; for example, transcriptomic libraries commonly use a 26- to 28-cycle cell-barcode and UMI read plus a 90- to 150-cycle transcript read, while ATAC libraries require paired-end reads. Recommended depth generally ranges from 20,000 to 50,000 read pairs per cell for whole-transcriptome profiling, with shallower depths acceptable for targeted panels. Sequencing runs include internal controls such as spike-in standards or well-characterized reference cell lines.
Bioinformatics Analysis and Reporting
Standard analysis pipelines perform raw read quality assessment, adapter trimming, reference-genome alignment, barcode correction, UMI deduplication, and matrix generation. Cells are filtered on metrics such as detected gene count, UMI count, and mitochondrial RNA fraction. Clustering and marker-gene identification support cell-type annotation, followed by subpopulation analysis, pseudo-time trajectory reconstruction, cell-cell communication inference, or clonotype tracking where applicable. Final deliverables include raw sequencing data (FASTQ), count matrices, analysis reports with publication-quality figures, and complete method documentation consistent with the laboratory's quality management system.
Applications
Single-cell sorting and sequencing experiments support a broad range of research, translational, and industrial objectives:
- Tumor heterogeneity and clonal evolution: resolving malignant subclones, rare resistant populations, and tumor microenvironment composition in cancers of the blood and solid organs.
- Immunology and vaccine research: profiling immune cell states, activation trajectories, and antigen-specific T and B cell responses following infection, vaccination, or immunotherapy.
- Developmental biology and stem cell engineering: reconstructing differentiation hierarchies, verifying cell fate transitions, and quality-controlling differentiated products intended for cell therapy.
- Drug discovery and pharmacology: identifying compound-responsive cell subsets, mechanism-of-action signatures, and toxicity-relevant cell states in preclinical models.
- Transplantation and autoimmune disease: monitoring immune reconstitution, detecting pathogenic effector clones, and tracking donor-recipient cell dynamics.
- Reference atlas construction: contributing to comprehensive cell atlases of human and model-organism tissues that serve as community baselines for future studies.
- Plant science and environmental microbiology: characterizing cell-type-specific regulation in crops and resolving population structure in microbial communities.
FAQ
What sample quality is required for a successful single-cell experiment?
Samples should be delivered as viable single-cell suspensions with viability above 80 percent (ideally above 85 percent for droplet-based platforms), minimal aggregates, and no visible debris. Fresh tissues should reach the laboratory within 24 to 48 hours in appropriate preservation medium; frozen PBMCs are acceptable when post-thaw viability meets the threshold.
How many cells can be captured, and how deeply are they sequenced?
Droplet-based platforms routinely capture 5,000 to 10,000 cells per channel and can be scaled to hundreds of thousands of cells across channels. Plate-based FACS sorting typically handles hundreds to a few thousand cells. Whole-transcriptome libraries are generally sequenced at 20,000 to 50,000 read pairs per cell, adjusted upward for rare-population discovery or deeper isoform coverage.
Can both fresh and frozen samples be analyzed?
Yes. Fresh material is preferred for fragile primary tissues because dissociation immediately after collection preserves cell states. Cryopreserved PBMCs and cell lines perform reliably when frozen, banked, and thawed under controlled conditions aligned with ISO 20387 biobanking guidance. Direct nuclei isolation from frozen tissue is an alternative when intact cytoplasm is not required, and is widely used for transcriptomic and ATAC profiling.
What controls ensure data reliability?
Each experiment includes cell-suspension QC, microfluidic or sorter calibration records, positive-control reference cells processed alongside the study sample, and post-run sequencing QC covering barcode knee plots, median genes per cell, doublet-rate estimates, and mapping statistics. Reports document all acceptance criteria so results are auditable and reproducible.
What bioinformatics deliverables are provided?
Deliverables typically include raw FASTQ files, filtered count matrices, QC summary tables, dimensionality-reduction and clustering figures, annotated cell-type compositions, marker-gene tables, and optional advanced analyses such as trajectory inference, differential abundance testing, and ligand-receptor interaction analysis, together with software versions and parameter records.
How long does a typical project take?
From sample receipt to finalized analysis report, standard projects commonly require two to four weeks, depending on sample type, number of cells, sequencing depth, panel complexity, and the scope of computational analysis. Feasibility assessment and protocol customization for challenging tissues may extend the timeline.