scplotkit

A pile of plotting functions for single-cell analysis, built on top of scanpy and AnnData, bundled up so they don’t have to be rewritten for every new project.

Masked and highlighted embeddings, gene co-expression blends, compositional bar plots and heatmaps, dataset overview panels, marker-gene matrix/violin plots, Sankeys/sunbursts/treemaps over annotation hierarchies — all styled consistently through a single PlotConfig and saved to disk with one call. Nothing statistical, no new analysis, just the plots. PRs for your own favorite single-cell (or spatial) plots are very welcome.

Contents

• Installation
  • From PyPI
  • From source
  • Requirements
  • Verifying the install
• Embeddings
  • Load the atlas
  • Set up the plotter
  • masked_umap
  • masked_umap_highlight
  • gene_expression_umap
  • gene_coexpression_umap
  • embedding_density
  • Next steps
• Composition
  • Load the atlas
  • Set up the plotter
  • stacked_barplots
  • stacked_barplots_multi_meta
  • composition_heatmap
  • abundance_bubble_grid
  • Next steps
• Overview
  • Load the atlas
  • Set up the plotter
  • sample_and_cell_counts_barplot
  • cells_per_patient_boxplot
  • cell_abundance_barplot
  • compare_cell_abundance_barplot
  • compare_cell_abundance_boxplot
  • Next steps
• Markers & expression
  • Load the atlas
  • Set up the plotter
  • rank_genes_matrix_and_dot
  • annotation_marker_matrixplot & annotation_marker_stacked_violin
  • pseudobulk_boxplot
  • pseudobulk_multigene_boxplot
  • ridgeline_plot
  • Next steps
• Hierarchy & enrichment
  • Load the atlas
  • Set up the plotter
  • sankey_plot
  • sunburst_plot & treemap_plot
  • ora_dotplot
  • Next steps
• API Reference
  • Configuration
  • ScPlotter (convenience class)
  • scplotkit.embeddings
  • scplotkit.composition
  • scplotkit.overview
  • scplotkit.markers
  • scplotkit.pseudobulk
  • scplotkit.ridgeline
  • scplotkit.enrichment
  • scplotkit.sankey

At a glance

import scanpy as sc
import scplotkit as spk

adata = sc.read_h5ad("my_atlas.h5ad")
plotter = spk.ScPlotter(output_dir="figures")

plotter.masked_umap(adata, color_by="cell_type", mask_values=["CD4 T", "CD8 T"], figure_name="Lymphocytes")
plotter.stacked_barplots(adata, level_column="cell_type", sample_column="sample_id")

See the Embeddings and the four notebooks that follow it for a walkthrough and demonstration of major plotting functions, or the API Reference for the complete function reference.