Single-Cell Data Analysis

Single-cell data is easy to generate and easy to over-interpret. We do the quality control properly, annotate cells against reference atlases, and tell you which clusters are real.

Single-cell experiments are expensive, and the analysis decisions made in the first week determine what the data can show. Filtering thresholds, integration method and clustering resolution are all judgement calls, and the wrong ones produce clusters that do not exist.

We document every one of those decisions and show you the effect of the alternatives, so the clusters in your paper are ones you can defend.

Data We Accept

  • 10x Genomics CellRanger output (filtered or raw matrices)
  • Raw FASTQ files for 10x, Drop-seq, Smart-seq2 or BD Rhapsody
  • Seurat, Scanpy/AnnData or SingleCellExperiment objects
  • scATAC-Seq fragment files and peak matrices
  • CITE-Seq antibody capture and hashtag data

Questions We Answer

  • What cell types are present, and in what proportions?
  • Which cell type is driving the difference between my conditions?
  • Are my clusters real biology or a technical artefact?
  • How do cell proportions shift across disease states or timepoints?
  • Which cells are signalling to which?

What We Do

Each project uses the subset of these that your research question requires.

01

Quality Control and Filtering

Empty droplet detection, ambient RNA correction, doublet identification, and filtering on genes per cell, counts and mitochondrial fraction — with the thresholds justified rather than copied from a tutorial.

02

Integration and Batch Correction

Merging samples, donors or timepoints using Harmony, Seurat CCA/RPCA or scVI, with quantitative assessment of whether biological signal survived the correction.

03

Clustering and Cell Type Annotation

Dimensionality reduction, graph-based clustering across a range of resolutions, and annotation using both marker-based and reference-based methods (SingleR, Azimuth, CellTypist) against relevant atlases.

04

Differential Expression and Composition

Pseudobulk differential expression between conditions within each cell type — the statistically correct approach — plus cell proportion testing across groups.

05

Trajectory and Cell Communication

Pseudotime and RNA velocity where the biology supports it, and ligand-receptor interaction analysis with CellChat or CellPhoneDB.

What You Receive

  • Annotated Seurat or AnnData object you can explore yourself
  • UMAP and cluster figures, coloured by sample, condition and cell type
  • Marker gene tables for every cluster with statistical support
  • Pseudobulk differential expression results per cell type
  • Cell proportion tables and composition test results
  • Analysis report documenting every filtering and clustering decision

Tools We Use

  • CellRanger, STARsolo, alevin-fry
  • Seurat, Scanpy, Bioconductor
  • SoupX, CellBender, DoubletFinder, scDblFinder
  • Harmony, scVI, scANVI
  • SingleR, Azimuth, CellTypist
  • Monocle3, scVelo, CellChat

Typical turnaround: 3–6 weeks depending on sample number and integration complexity

Indicative price: From $1,200 for a single sample through to annotation

Reduced rates are available for students and researchers at African public institutions. Every project is quoted in writing before work begins.

Services are provided for research purposes only. They are not intended for clinical diagnosis, treatment decisions or individual health assessment. See how it works, data submission guidelines and what you receive.

Contact DataCore Analytics

Tell us about your data and we will scope it — free, within one working day.

+233 558 017 827