Sample analysis reports

See what a Bridge Horizon report looks like

These are real outputs from our single-cell RNA-seq and differential expression pipelines. Every plot is publication-ready, downloadable for your slides and papers, and backed by a fully traceable run.

Report 1

Single-cell RNA-seq summary report

This is the summary report every single-cell run produces on Bridge Horizon. It walks from quality control through integration, clustering, doublet detection, and cell-type annotation, with each plot rendered at publication quality.

In the interactive version, your team clicks any plot to download a static image for presentations and publications, and every figure links back to the exact run, parameters, and pipeline version that produced it.

UMAP of single-cell RNA-seq data coloured by Seurat clusters at the analysis resolution, the reference map for all downstream annotation — The reference UMAP coloured by Seurat clusters. Every downstream annotation, marker table, and cell-type label is placed on this map.

UMAP coloured by automatic cell-type labels assigned from marker-gene module scores in a single-cell RNA-seq report — Automatic cell-type labels from marker-gene module scores, projected onto the same UMAP so bench scientists and bioinformaticians look at identical results.

Violin plots of per-cell QC metrics (genes detected, UMI counts, mitochondrial fraction) for one single-cell sample — Per-cell QC metrics for every sample: genes detected, UMI counts, and mitochondrial fraction, with the filtering cutoffs recorded on the run.

Grid of UMAP embeddings of integrated single cells at multiple clustering resolutions — UMAP embeddings across clustering resolutions, used to pick the granularity that separates real biological populations without over-splitting.

Doublet scores projected on the single-cell UMAP, flagging likely doublets before annotation — Doublet scores projected on the UMAP. Likely doublets are flagged and removed before annotation so they never inflate marker signal.

Stacked bar chart of each cluster's proportion within every sample, showing cell-type composition changes — Cluster proportions per sample: the primary view of how cell-type composition shifts between conditions.

Cluster UMAP split into one panel per sample, revealing cluster presence and density shifts between samples — The same cluster UMAP split per sample, revealing cluster presence, absence, and density shifts at a glance.

Stacked bar chart of cell-type composition per sample using marker-based labels — Cell-type composition per sample using marker-based labels, showing how populations shift before any manual annotation.

Dot plot of a curated marker-gene panel across all single-cell clusters for confirming cluster identities — Marker-gene dot plots across all clusters, used to confirm cluster identities against curated gene lists (exhaustion, naive/memory, Treg, effector).

Two-panel figure: UMAP coloured by final cell annotation beside a 2D cell-density map of the same embedding — Final annotation beside a 2D cell-density map of the same UMAP, highlighting where cells concentrate and how densities shift between samples.

Report 2

Differential expression report: condition 1 vs condition 2

When you compare conditions, Bridge Horizon runs DESeq2, edgeR, and limma-voom on the same counts and reports where they agree. Instead of betting your conclusions on a single method's assumptions, you see the consensus, and exactly where the methods disagree.

Every figure is downloadable, and the full run is reproducible: design formula, contrasts, cutoffs, and package versions are all captured automatically.

DESeq2 volcano plot of log2 fold change against adjusted p-value for condition 1 versus condition 2 — DESeq2 volcano plot: log2 fold change against adjusted p-value, with significantly up- and down-regulated genes coloured.

DESeq2 MA plot of log2 fold change against mean normalised expression with significant genes highlighted — DESeq2 MA plot. A cloud centred on zero at every expression level confirms normalisation behaved; significant genes are highlighted.

PCA of variance-stabilised RNA-seq counts, one point per sample, checking that conditions separate — PCA on variance-stabilised counts. Conditions should separate on at least one axis; a mislabelled or outlier sample is caught here, before anything downstream is trusted.

Heatmap of top differentially expressed genes, z-scored per gene, clustered across samples — Heatmap of the top differentially expressed genes, clustered on both axes. Samples should group by condition.

Venn-style overlap of significant gene sets between DESeq2, edgeR, and limma-voom — Overlap of significant gene sets across DESeq2, edgeR, and limma-voom. The centre is the high-confidence set to take forward.

UpSet plot of significant-gene set sizes for every combination of differential expression methods — UpSet view of every method combination, so you can see at a glance how much the methods agree on significance.

Need a report like this on your own data?

Run it yourself on Bridge Horizon, or have our team produce it as a service. Either way you get the same traceable, publication-ready output your whole team can explore together.

Single-cell RNA-seq on Bridge Horizon

How the Seurat workflow behind Report 1 runs end to end, from count matrix to annotated clusters.

Read the guide

DESeq2 differential expression guide

The practical guide behind Report 2: inputs, design formulas, contrasts, and reproducible runs.

Read the guide

See how Horizon fits your team.

Built on production genomics experience since 2020, Horizon makes every dataset, pipeline run, and parameter shared and traceable across your entire team.