Cluster Markers — Why a Cluster Is Named That

Header: 🔎 Cluster Markers

Available on single-cell RNA-seq projects.

The Cluster Markers tab answers: why is this cluster called what it is called? Every cell type label elsewhere in the project comes from here, and this tab shows the evidence behind each one so you can accept it, question it, or relabel it in your own write-up.

Example output from the sample project.

Cluster Annotation

An expandable table lists every cluster and how it was named:

Column	Meaning
cluster	cluster number
label	the cell type name used throughout the project
confidence	the strength of the evidence
method	how the label was assigned

Two methods appear, and they report confidence differently. SingleR compares each cluster against a reference dataset and reports a percentage. marker_table matches the cluster's top genes against a marker list and reports how many matched — so a confidence of "1 marker hits" means the label rests on a single gene. Treat those labels as provisional and check them against the marker plot below.

Cluster

The Cluster dropdown selects which cluster to inspect. Each entry shows the label, cluster number, method, and confidence together — for example, "Chondrocyte_c0 (cluster 0, marker_table, 1 marker hits)" — so the weakest labels are visible while you choose.

Top Markers

The bar chart shows the selected cluster's strongest marker genes ranked by avg log2FC — how much higher each gene is in this cluster than in the rest of the cells. These are the genes that define the cluster, and the ones to recognise (or fail to recognise) when judging a label.

The number of genes shown follows Top N in the sidebar Filters section.

Search the Full Marker Table

A checkbox, "Search the full marker table (not just top N)", extends the table beneath the chart to every marker gene rather than the top-ranked few. Use it to ask whether a specific gene you expect in a cell type is present at all — a gene can be a genuine marker without ranking in the top few by fold change.