Single-Cell Sequencing Technologies

Introduction to Single-Cell Biology

Single-cell sequencing technologies measure molecular profiles of individual cells rather than averaging across populations, revealing heterogeneity invisible in bulk assays. Even genetically identical cells in the same tissue occupy diverse transcriptional, epigenetic, and proteomic states—varying by cell type, cell cycle phase, metabolic state, microenvironmental position, and stochastic gene expression fluctuations. Bulk RNA-seq averages these differences obscuring cell type compositions, rare cell populations, and state-specific regulatory programs. Single-cell technologies have transformed our understanding of tissue organisation, developmental processes, immune heterogeneity, and tumour cell states.

The first single-cell RNA-seq protocol was published by Tang et al. in 2009 profiling a single mouse blastomere. The field transformed with microfluidic droplet and nanowells enabling 10,000s of cells per experiment at falling costs. The Human Cell Atlas initiative is systematically mapping all human cell types using single-cell multi-omics across all tissues. Disease applications identify cancer subpopulations driving resistance, tumour-infiltrating immune cell states correlating with immunotherapy response, and neurodegenerative disease-specific neuronal vulnerability signatures. Single-cell sequencing is arguably the most impactful technology advance in cell biology since the fluorescence microscope.

Single-Cell RNA Sequencing

Droplet-Based Methods

10x Genomics Chromium and similar droplet microfluidic platforms encapsulate individual cells with barcoded beads in aqueous droplets suspended in oil. Cell lysis in the droplet releases mRNA which hybridises to poly-dT primers on the bead, reverse-transcribed with cell-barcode and UMI (unique molecular identifier) sequences. After pooled library amplification and sequencing, cell barcodes demultiplex reads to individual cells; UMIs deduplicating PCR copies enable quantitative counting of mRNA molecules. Typical experiments capture 1000-10,000 genes per cell across 5000-50,000 cells—sufficient to identify dozens of cell types by clustering transcriptional similarity. Cost has fallen below $1 per cell at scale, enabling population-scale studies.

Analysis of scRNA-Seq Data

scRNA-seq analysis workflows typically involve: quality filtering (removing low-quality cells and empty droplets), normalisation (correcting sequencing depth variation), dimensionality reduction (PCA followed by UMAP or t-SNE for visualisation), clustering (Leiden or Louvain graph-based algorithms), and cell type annotation (marker gene expression, automated label transfer from reference atlases). Trajectory analysis (pseudotime, RNA velocity using spliced/unspliced mRNA ratios) infers developmental or differentiation trajectories ordering cells along continuous state transitions. Differential expression between conditions identifies genes dysregulated in disease. Cell-cell communication inference uses receptor-ligand co-expression pairs to predict intercellular signalling in tissue contexts.

Single-Cell Multi-Omics

Chromatin Accessibility and Multi-Omics

Single-cell ATAC-seq (scATAC-seq) profiles chromatin accessibility using Tn5 transposase in individual cells—identifying open regulatory elements (promoters, enhancers) cell-by-cell. Cell type-specific accessibility landscapes define regulatory vocabularies: each cell type has distinct sets of open enhancers binding its combination of transcription factors. Simultaneous measurement of chromatin accessibility and gene expression from the same cells (10x Multiome, SHARE-seq) links regulatory element activity to target gene expression, enabling cis-regulatory element-to-gene mapping. Single-cell ATAC-seq of tumours identifies cancer-specific regulatory programs and reveals transcription factor activity alterations driving oncogenic states.

Spatial Transcriptomics

Spatial transcriptomics maps gene expression while preserving cellular spatial position information within tissue sections. Visium (10x Genomics) captures mRNA from 55 micron spots on tissue sections (~1-10 cells per spot) using spatially barcoded oligonucleotide arrays. Slide-seq and Stereo-seq achieve near-single-cell spatial resolution. MERFISH and seqFISH use sequential RNA FISH rounds to multiplex hundreds of genes with single-molecule resolution while preserving spatial information. Spatial transcriptomics reveals tissue zonation (liver zone-specific gene expression), tumour microenvironment spatial organisation (immune cell distribution relative to tumour core), and ligand-receptor interactions between spatially adjacent cells—physical proximity context invisible without spatial information.

Applications in Medicine and Biology

The Human Cell Atlas consortium is creating a comprehensive reference map of all human cell types using single-cell multi-omics across 30+ organs in healthy donors spanning developmental stages and demographic diversity. Initial atlases of lung, liver, placenta, heart, eye, gut, brain, and immune system have been published, discovering previously unknown cell types (Ionocytes in the airway, fat-associated lymphoid clusters), rare differentiated states, and new markers. These reference maps accelerate disease interpretation—finding which cell types express disease GWAS genes to translate genetic discoveries into mechanistic cell biological insights, identifying cellular targets for gene therapy, and providing normal references for comparison in disease studies.

Examples and Applications

Example 1: Cancer Cell State Heterogeneity

scRNA-seq of glioblastoma revealed that individual tumours contain cells traversing a continuous spectrum of differentiation states including stem-like, mesenchymal, astrocyte-like, and oligodendrocyte progenitor-like states. This continuous cell state landscape (cellular plasticity) reframes therapy resistance: cells in different states differ in drug sensitivity; killing cells in one state allows others to repopulate tumours. scRNA-seq identified the NPC-like (neural progenitor) state as the most drug-resistant in GBM. Anti-tumour therapy must target all states or prevent state transitions, requiring understanding of state transition regulators—identified by trajectory analysis and regulon inference from scRNA-seq data.

Example 2: Tumour-Infiltrating Immune Cells

scRNA-seq of tumour-infiltrating lymphocytes (TILs) revealed that T cell dysfunction (exhaustion) in tumours exists along a developmental spectrum from progenitor exhausted T cells (Tpex, TCF7+) through terminally exhausted (Tim-3+, Tox+). Anti-PD-1 immunotherapy preferentially expands Tpex cells—whose abundance predicts response—rather than reinvigorating terminally exhausted cells. This understanding explains why tumours with minimal Tpex cells respond poorly to PD-1 blockade. CXCL13-expressing CD8 T cells in the vicinity of tumour cells predict immunotherapy benefit. scRNA-seq thus mechanistically explains clinical response biomarkers and guides development of combination strategies expanding Tpex infiltration.

Example 3: Atlas of Human Embryonic Development

Single-cell atlases of human embryonic development at days 16-70 (Carnegie stages 1-14) revealed previously invisible cell populations and differentiation trajectories. Spatial transcriptomics of gastruloids and actual embryos mapped cell migration routes, signalling gradients, and lineage specifications. These data reveal how the three germ layers form, how neural crest cells delaminate and migrate, and how organ primordia develop—providing developmental ground truth for assessing fidelity of iPSC-derived organoid models. Developmental atlases also identify when disease-causing genes are expressed in susceptible progenitors, informing windows of therapeutic intervention and mechanisms of congenital malformations.

Example 4: Peripheral Blood Immune Profiling

scRNA-seq of peripheral blood mononuclear cells (PBMCs) enables comprehensive immune profiling—mapping all circulating immune cell types and states simultaneously. CITE-seq adding antibody-based protein measurement alongside transcriptomics provides multi-modal immune phenotyping. Population atlases of PBMCs in thousands of healthy donors across age, sex, and ancestry characterise normal immune variation. Disease applications include COVID-19 immunopathology characterisation (neutrophil and monocyte emergency granulopoiesis correlating with severity), rheumatoid arthritis synovial and blood immune composition, SLE flare signatures, and checkpoint immunotherapy toxicity mechanisms—all identified by scRNA-seq cell state and proportion changes relative to healthy reference atlases.

Example 5: Spatial Transcriptomics in Brain

MERFISH spatial transcriptomics in mouse brain mapped 1100 genes at single-cell resolution across brain regions, identifying 300+ cell types with molecularly distinct spatial distributions. The data revealed fine-scale spatial organisation of cell types within brain regions at sub-layer resolution, and identified new cell types restricted to specific microdomains. Human hippocampus spatial transcriptomics in Alzheimer's disease showed entorhinal cortex layer II stellate cells—the cells most vulnerable to early neurofibrillary tangles—have distinctive molecular signatures identifiable by spatial transcriptomics enabling study of their selective vulnerability. Spatial transcriptomics of schizophrenia and ASD brains is identifying layer-specific gene expression changes.

Example 6: Lineage Tracing with Single-Cell Barcoding

CRISPR recording and DNA barcoding enable lineage tracing at single-cell resolution—recording the history of past cell divisions in the genome. GESTALT/scGESTALT uses CRISPR diversification of a genomic target array generating a unique barcode read by scRNA-seq, recording clonal history alongside transcriptional identity. TracerSeq labels progenitors with heritable integrations sequenced alongside transcriptomes. These approaches reconstructed zebrafish embryonic lineage trees from a single cell, enabling identification of which progenitor cells give rise to which organs at which developmental stages. Combining lineage information with transcriptional state enables reconstruction of complete developmental trajectories of individual cells during organogenesis.

Example 7: Clonal Haematopoiesis Analysis

Clonal haematopoiesis of indeterminate potential (CHIP) occurs in ~10% of people over 65 with DNMT3A, TET2, ASXL1, or JAK2 mutations in HSC clones. scRNA-seq of clonal haematopoiesis samples identified inflammatory gene expression programs in TET2-mutant monocytes and macrophages—explaining why CHIP associates with 2-fold increased cardiovascular disease risk through accelerated atherosclerosis. IL-6 and IL-8 inflammatory pathway activation in TET2-mutant clones can be suppressed with IL-6 inhibitors. Understanding clonal haematopoiesis through single-cell precision is guiding clinical trials testing whether anti-inflammatory intervention in CHIP carriers reduces cardiovascular risk.

Example 8: Single-Cell Proteomics by Mass Spectrometry

SCoPE-MS and related single-cell proteomics methods use multiplexed tandem mass tag (TMT) labelling of individual cells to profile 1500-3000 proteins per cell. Unlike antibody-based protein measurement (limited to ~100 targets), MS-based single-cell proteomics is comprehensive and unbiased. Current throughput (hundreds to low thousands of cells) lags scRNA-seq by one to two orders of magnitude, limiting statistical power for rare cell populations. Technological advances in nanowell sample preparation, low-injection-volume LC, and next-generation mass spectrometers promise to bring single-cell proteomics to the scale of scRNA-seq. Establishing proteome-transcriptome correlation cell-by-cell will reveal post-transcriptional regulation at single-cell resolution.

Try it live

Everything above runs in your browser — open Single-Cell Sequencing Simulator and change the parameters while it is running. Nothing is installed, nothing is uploaded, the whole model lives in one tab.