✕
CN EN

Bulk RNA-seq analysis

For tissue, cell line, blood and other routine samples: two-group or multi-group comparisons, time series, treatment versus control. The comparison design is written into the analysis plan before any work starts.

Standard analysis

  • Raw read QC and trimming (FastQC, fastp)
  • Alignment and quantification (HISAT2/STAR + featureCounts, or Salmon)
  • Sample correlation, PCA and clustering to spot outliers and batch effects
  • Differential expression (DESeq2/edgeR) with volcano, heatmap and MA plots
  • GO / KEGG enrichment with bubble plots
  • Expression tables (counts, TPM/FPKM)

Optional advanced modules

  • GSEA
  • WGCNA co-expression modules and trait association
  • Alternative splicing (rMATS)
  • Fusion gene detection
  • SNP/InDel calling from RNA-seq
  • Transcription factor and pathway activity inference
  • Model-based analysis for time series and multi-factor designs

What you send and what comes back

You provideYou receive
fastq files (or an existing expression matrix)QC report and alignment/quantification statistics
Sample sheet: groups, batches, replicatesNormalised expression matrix and differential gene tables (Excel/CSV)
Species and reference genome version (we can advise)Publication-ready figures: PCA, volcano, heatmap, enrichment (PDF/SVG/PNG)
The question you want answered and the target journal (optional)Analysis code and parameters, plus a methods paragraph

Typical turnaround: 5–10 working days for the standard scope; complex designs as stated in the written analysis plan. Quotes are per project, returned within 1 working day.

Sample deliverable (real run on public data)

Volcano plot
Volcano plot
Pathway enrichment
Pathway enrichment

Open the full sample report

Can you run differential expression with only 2 replicates per group?

We can run it, but statistical power is limited; the report states the limitation and suggests validation. Three or more biological replicates per group is the usual requirement and we say so at the plan stage.

I already have an expression matrix from another provider. Can you do only the downstream analysis?

Yes. Send the matrix and the sample sheet; we first check whether it holds counts or TPM/FPKM and choose the method accordingly.

Which reference genome version do you use?

By default the current Ensembl or NCBI release, matched to any data you already have. The version is stated in the plan and in the report.

Single-cell RNA-seq analysis

For 10x Genomics, BD Rhapsody and other droplet- or microwell-based platforms, and for published public datasets. We start from fastq or Cell Ranger output, and also accept Seurat / AnnData objects.

Standard analysis

  • Cell and gene QC: mitochondrial fraction, doublet detection, thresholds derived from the data
  • Normalisation, variable genes, dimensionality reduction (PCA/UMAP/t-SNE)
  • Multi-sample integration and batch correction (Harmony, Seurat CCA/RPCA)
  • Clustering and cell-type annotation: reference-based annotation checked against canonical markers
  • Marker genes and differential expression per cell type
  • Cell-proportion changes between groups

Optional advanced modules

  • Pseudotime / trajectory (Monocle, PAGA, RNA velocity)
  • Cell–cell communication (CellChat, CellPhoneDB)
  • Regulatory networks (SCENIC)
  • CNV inference (inferCNV) for tumour-cell identification
  • Sub-clustering and functional enrichment
  • Single-cell / bulk integration (deconvolution, gene-set scoring)

What you send and what comes back

You provideYou receive
fastq or Cell Ranger output (filtered matrix)QC report with every threshold and its rationale
Sample sheet: groups, batches, tissueAnnotated object (Seurat .rds or .h5ad)
The cell types or questions you care aboutPublication-ready figures: UMAP, marker dot plots, proportion bars, heatmaps
Published reference annotations (optional)Differential gene and enrichment tables per cell type, code and methods paragraph

Typical turnaround: 10–15 working days for the standard scope; complex designs as stated in the written analysis plan. Quotes are per project, returned within 1 working day.

Sample deliverable (real run on public data)

UMAP with cell-type labels
UMAP with cell-type labels
Marker gene dot plot
Marker gene dot plot

Open the full sample report

Only 1–2 samples per group — can you still test differences?

We can produce results, but cell-level tests overstate significance. Where sample numbers allow we aggregate to sample level (pseudobulk); otherwise the report states the limitation.

How is annotation kept reliable?

Reference-based labels are checked against canonical markers. Clusters that markers do not support keep their number rather than a name, and the evidence is listed in the report.

Will integration erase the biology?

After integration we check that the treatment-versus-control differences are still present and show before/after comparisons in the report.

Spatial transcriptomics analysis

For 10x Visium, Visium HD, Xenium, MERFISH and other spatial platforms. We start from Space Ranger or platform output and return results registered to the tissue image.

Standard analysis

  • Spot / cell QC and filtering
  • Normalisation, dimensionality reduction, clustering and spatial domains
  • Spatially variable genes (SVGs)
  • Expression and cluster overlays aligned to the tissue image
  • Spatial distribution of known marker genes
  • Comparison across sections

Optional advanced modules

  • Cell-type deconvolution with a single-cell reference (cell2location, RCTD)
  • Neighbourhood analysis and cell-type co-localisation
  • Spatial ligand–receptor interaction
  • Manually annotated regions (e.g. tumour/stroma) and regional differential analysis
  • Integration of single-cell and spatial data

What you send and what comes back

You provideYou receive
Space Ranger or platform output including the tissue imageQC report and spatial domain results
Section sheet: groups, tissue type, pathology annotation if anyPublication-ready overlays on the tissue image (PDF/SVG/PNG)
Matched single-cell reference (optional, for deconvolution)SVG and regional differential tables

Typical turnaround: 10–15 working days for the standard scope; complex designs as stated in the written analysis plan. Quotes are per project, returned within 1 working day.

No public example is available for this analysis yet. Deliverables follow the same structure as our sample reports (report, figures, tables, code, methods text); the specific figures are listed in the written plan.

No matched single-cell data — can you still deconvolve?

Yes, using a public single-cell dataset from the same tissue as reference; the report states the source and its limits.

Can regions drawn by a pathologist be used?

Yes. Send the annotated image or coordinates and we compare by region.

Non-coding RNA analysis (lncRNA / circRNA / miRNA)

For strand-specific total RNA and small RNA sequencing. Beyond identification and quantification, the emphasis is on linking non-coding RNAs to target genes and pathways.

Standard analysis

  • lncRNA identification (coding potential), quantification and differential analysis
  • circRNA detection (intersection of CIRI2, find_circ and others) and quantification
  • miRNA identification (miRDeep2), known and novel miRNA quantification and differential analysis
  • Target prediction and enrichment
  • Expression pattern clustering and heatmaps

Optional advanced modules

  • ceRNA network (lncRNA/circRNA–miRNA–mRNA) construction and visualisation
  • lncRNA cis / trans target analysis
  • Integration with mRNA data from the same samples
  • Conservation and structural annotation of candidates

What you send and what comes back

You provideYou receive
Total RNA or small RNA fastq filesIdentification and expression tables for each RNA class
Sample sheet and comparison designDifferential results with enrichment figures
Species and reference genome versionceRNA network files (Cytoscape-ready)

Typical turnaround: 7–12 working days for the standard scope; complex designs as stated in the written analysis plan. Quotes are per project, returned within 1 working day.

No public example is available for this analysis yet. Deliverables follow the same structure as our sample reports (report, figures, tables, code, methods text); the specific figures are listed in the written plan.

Why do circRNA results differ so much between tools?

Tools use different criteria for back-splice junctions. We report the intersection of several tools as the high-confidence set and also provide each tool's full list.

My species has no miRNA annotation.

We use the miRBase annotation of a close relative plus novel miRNA prediction, and state the basis in the report.

Tell us the data type and the question you want answered. You will receive a written analysis plan and quote within 1 working day.

Request an analysis plan