-
Notifications
You must be signed in to change notification settings - Fork 0
Tab Cell Coloring
srao edited this page Aug 17, 2026
·
3 revisions
The Coloring tab controls how cells in the segmentation layer are rendered: you can colour them by continuous gene expression or by discrete cluster assignment, and optionally hide clusters that are too small or not of interest. Changes take effect only when you click Apply Cell Coloring, which also updates the linked UMAP window.

| Control | Description |
|---|---|
| White background | Checkbox. Toggles the napari canvas background between black (default) and white. |
| Color cells by | Radio buttons: Gene Expression or Cluster. Selects the colouring mode. |
| Gene | Dropdown (all gene names). Gene to display when Gene Expression mode is active. |
| Colormap | Dropdown. Colour map applied to gene expression values (default: viridis). Options include viridis, plasma, magma, inferno, cividis, and others. |
| Clustering | Dropdown. Clustering used for Cluster mode and for cluster-level visibility filtering. |
| Filter by cluster | Checkbox. When enabled, reveals a grid of per-cluster checkboxes for manual visibility control. |
| Min cluster size | Slider (100–10 000, default 500, in steps of 100). Cell-count threshold used by Filter Small Clusters. |
| Filter Small Clusters | Button. Unchecks clusters whose cell count is below Min cluster size. Also ticks Filter by cluster for you, so you do not need to enable it first. |
| Edit Cluster Labels... | Button. Opens a dialog to rename the clusters of the selected clustering. The same dialog is available in the Rank Genes tab. |
| Select All | Button. Checks all per-cluster visibility checkboxes. |
| Deselect All | Button. Unchecks all per-cluster visibility checkboxes. |
| Per-cluster checkboxes | Grid. Shows one checkbox per cluster; uncheck a cluster to hide those cells. |
| Apply Cell Coloring | Button. Applies all current settings to the cell labels layer and refreshes the UMAP window. |
- Choose a Clustering from the dropdown if you intend to colour by cluster or filter by cluster identity.
- Select your colouring mode with the Color cells by radio buttons.
- In Gene Expression mode, pick a Gene and Colormap.
- In Cluster mode, each cluster receives a distinct categorical colour.
- To hide unwanted clusters, enable Filter by cluster, then either use Filter Small Clusters with an appropriate Min cluster size, or uncheck individual clusters manually.
- Click Apply Cell Coloring to commit changes to the viewer.
- The display does not update in real time; you must click Apply Cell Coloring after every change.
- Cluster filtering and gene expression colouring are fully compatible: you can colour cells by gene expression while hiding specific clusters.
- Colouring settings are not persisted. Every launch starts in Gene Expression mode on the first gene in the panel with the viridis colourmap and no cluster filter, whatever you had applied when you closed the viewer. The clusterings and cluster names themselves are saved; the way you were looking at them is not.
- The cluster filter set here is read by several other tabs — Nhood Enrich, Co-occur, Lig-Rec and parts of ROI DEG restrict their results to the clusters left visible here. If an analysis returns fewer groups than expected, check whether a filter is still active.
- Cells can also be coloured by a continuous per-cell score rather than a gene — the CNV tab uses this for CNV burden, and the Annot Dist tab for distance.
Reference
Cells
Genes
Spatial
- ROI Analysis
- Ligand-Receptor
- Neighborhood Enrichment
- Co-occurrence
- Spatial Domains
- Annot Nhood
- Annot Distance
Images
Tools
Tutorials
- Getting Started
- Clustering and DEG
- H&E Registration
- ARMS Overlay
- ROI Analysis
- Annotations
- Recovering a Cache