Visualizing spatially resolved biological data with appropriate color mapping can significantly facilitate the exploration of underlying patterns and heterogeneity. Spaco (spatial colorization) provides a spatially constrained approach that generates discriminate color assignments for visualizing single-cell spatial data in various scenarios.
Color assignment
By quantifying the complex topology between cell type clusters, We optimized color assignment of to achieve better visual recognizability.
Palette extraction
We provide a method for extracting color plates from images. While maintaining the theme color, the color differentiation is maximized.
# Latest source from github (Recommended)
pip install git+https://github.com/BrainStOrmics/Spaco.git
# PyPI
pip install spaco-release
- python>=3.8.0
- numpy>=1.18.0
- pandas>=0.25.1
- scipy>=1.10.0
- anndata>=0.8.0
- scikit-learn>=0.19.0
- scikit-image>=0.19.0
- colormath>=3.0.0
- pyciede2000==0.0.21
- umap-learn>=0.5.0
- logging-release>=0.0.4
- typing_extensions>=4.0.0
NOTE THAT: Currently we found numpy version (1.22.x or 1.23.x) could influence the result of graph-guided mode of Spaco; However, colorization should be acceptable with either version; To exactly reproduce the results in Spaco vignette or paper, please check the numpy version at the end of each jupyter notebook.
import spaco
import scanpy as sc # For visualization
import squidpy as sq # For loading example dataset
# loading data
adata_cell = sq.datasets.seqfish()
palette_default = adata_cell.uns['celltype_mapped_refined_colors'].copy()
# color assignment with default palette
color_mapping = spaco.colorize(
cell_coordinates=adata_cell.obsm['spatial'],
cell_labels=adata_cell.obs['celltype_mapped_refined'],
palette=palette_default,
radius=0.05,
n_neighbors=30,
)
# Order colors by categories in adata
color_mapping = {k: color_mapping[k] for k in adata_cell.obs['celltype_mapped_refined'].cat.categories}
palette_spaco = list(color_mapping.values())
# Spaco colorization
sc.pl.spatial(adata_cell, color="celltype_mapped_refined", spot_size=0.035, palette=palette_spaco)- A brief demo is included in Spaco package.
- Please see our STAR Protocol for detailed description
- Working with R? See SpacoR.
Scripts to reproduce benchmarking and analytic results in Spaco paper are in repository Spaco_scripts
Users can use issue tracker to report software/code related issues. For discussion of novel usage cases and user tips, contribution on Spaco performance optimization, please contact the authors via email.