Vitessce supports several components for visualization of single-cell data, and several controller components for updating visualization parameters. The component names found on this page are all valid values for the field
layout.component in the view config. On this page you will also find screenshots of each component.
The terms "component" and "view" are used interchangeably thoughout this documentation.
The scatterplot component displays 2-dimensional (pre-computed) embeddings / projections (such as t-SNE or UMAP).
Use with the
embeddingType coordination type to specify which embedding to use when mapping cells onto scatterplot points.
Use with datasets containing the
cells data type. Optionally, the
genes data types can be used for coloring cells on the plot.
The heatmap component displays a (normalized) cell-by-gene (or gene-by-cell) matrix visualization.
This component uses the
expression-matrix data type, and optionally displays cell set color assignments if a file with the
cell-sets data type is available in the same dataset.
The heatmap can support very large matrices (~6,000 x ~9,000) when using Zarr-based file types (best performance can be acheived when arrays are saved with the type
The spatial component is meant to display data with spatial coordinates, including spatially-resolved cell segmentations as polygons (from the
cells data type) and molecule positions as points (from the
molecules data type). The spatial component also includes a multiplexed and multi-scale image viewer (implemented with Viv). Image files can be defined with the
raster data type. Cell segmentations can be colored by gene expression data through the
expression-matrix data type or cell set color assignments through the
cell-sets data type.
The layer controller component provides an interface for manipulating the visualization layers displayed in the
The image layers are coordinated through the
spatialRasterLayers coordination type.
The genomic profiles component displays genome browser tracks (using the
genomic-profiles data type) containing bar plots, where the genome is along the x-axis and the value at each genome position is encoded with a bar along the y-axis.
Genome tracks may be colored by corresponding cell set color assignments if the
cell-sets data type is available in the same dataset, coordinated on the
cellSetSelection coordination type.
This component was implemented with HiGlass.
The genes component displays an interactive list of genes when used with datasets containing the
expression-matrix data type.
The cell sets component displays an interactive list of (potentially hierarchical) cell sets when used with datasets containing the
cell-sets data type. This can be useful for managing cell sets representing clustering algorithm outputs or cell type annotations.
The cell set sizes component displays a bar plot with the currently-selected cell sets (using the
cell-sets data type and
cellSetSelection coordination type) on the x-axis and bars representing their size (by number of cells) on the y-axis. This component was implemented with Vega-Lite.
The description component can be used to display text content. This may be useful for communicating the details of a dataset or a data analysis process. When the
spatial component is used for visualization of imaging data, the
description component also renders a dropdown containing image metadata.
The status component displays debugging messages, including app-wide error messages when datasets fail to load or when schemas fail to validate. Details about the entity under the mouse cursor (cell, gene, and/or molecule) are displayed during hover interactions.
The cell set expression component displays a violin plot with expression values for the selected gene, implemented with Vega-Lite.
The expression histogram component displays a bar plot with the distribution of expression values for the selected gene, implemented with Vega-Lite.