Single-Cell RNA-Seq (scRNA-Seq) Services

Single-cell RNA-Seq (scRNA-Seq) analysis

Single-cell RNA sequencing (scRNA-Seq) measures gene expression in individual cells, making it possible to dissect cellular heterogeneity, identify rare cell populations and map the diversity of tissues that traditional bulk RNA-Seq cannot resolve. It has transformed our understanding of development, immunology, cancer and many complex tissues.

We provide complete scRNA-Seq analysis using Seurat, Scanpy and related tools, from raw count matrices to fully annotated, biologically interpretable results.

Our Services

• Quality control, doublet detection and normalisation
• Dimensionality reduction, clustering and UMAP / t-SNE visualisation
• Cell-type annotation and marker-gene identification
• Differential expression between clusters and conditions
• Trajectory and pseudotime analysis
• Cell–cell communication and ligand–receptor analysis
• Batch-effect correction and multi-sample integration
• Gene-set and pathway-enrichment analysis

The result is a clear picture of which cell types are present and how their transcriptional programmes change across conditions, time and treatment.

What you receive

• A quality-controlled, normalised expression matrix with doublets and low-quality cells removed
• Annotated cell clusters, each supported by marker genes and reference-based evidence
• Differential expression results between clusters, conditions or timepoints
• Pathway and gene-set enrichment for each population of interest
• Trajectory, cell-communication or integration analyses where they fit the biological question
• Publication-ready UMAPs, dot plots, violin plots, heatmaps and proportion plots
• A written report and the full analysis code, so results are reproducible and extensible

How the analysis runs

• Quality control on genes per cell, UMI counts and mitochondrial fraction, with thresholds chosen from your data rather than copied defaults
• Doublet detection and removal with scDblFinder, DoubletFinder or Scrublet
• Normalisation and selection of highly variable genes
• Dimensionality reduction, neighbourhood graph construction and clustering, with resolution chosen against marker-gene coherence rather than left arbitrary
• Cell-type annotation combining marker genes, reference-based automated annotation and manual review
• Differential expression using pseudobulk methods where you have biological replicates, which control false positives far better than per-cell tests
• Downstream analyses: pseudotime, RNA velocity, ligand-receptor communication, gene-regulatory networks

Batch effects and integration

Multi-sample experiments almost always carry batch effects, and the risk runs both ways: too little correction leaves samples clustering by batch rather than biology, while too much erases the very differences you set out to measure.

We select the integration method to suit the design — Harmony, Seurat anchors, scVI or scANVI — and assess the result quantitatively rather than by eye, checking that batches mix while distinct cell types stay separate.

Getting differential expression right

Treating individual cells as independent replicates is the most common statistical error in single-cell work. Cells from one animal or donor are not independent samples, and per-cell tests on such data generate large numbers of false positives that do not replicate.

Where your design includes biological replicates, we aggregate to pseudobulk per sample per cell type and test with established bulk methods. Where it does not, we say so and interpret accordingly.

Data we accept

10x Genomics Cell Ranger output, raw FASTQ, count matrices in matrix, H5 or H5AD format, Seurat or AnnData objects, or public data from GEO, the Single Cell Portal and the Human Cell Atlas. Droplet, plate-based and nuclei protocols are all supported, as are CITE-Seq and hashed multiplexed designs.