sjanpy

Subjacent Analysis Toolkits for Single-Cell Omics in Python

A collection of visualization and analysis utilities designed to enhance single-cell RNA-seq workflows. Built on top of Scanpy and AnnData, sjanpy provides publication-ready plotting functions and efficient analysis tools.

Features

• Nebulosa Density Plots - Weighted kernel density estimation to address the overplotting problem in single-cell visualizations
• Differential Expression Analysis - Fast vectorized DEG computation with volcano plots and cluster-level comparisons
• Advanced Dot Plots - Hierarchical clustering, K-means grouping, dendrograms, and fan-shaped polar layouts
• Enhanced Embeddings - High-quality UMAP/t-SNE visualizations with density overlays and smart labeling
• Stacked Bar Plots - Cell composition analysis with intelligent label placement
• Gene Filtering - Utilities to remove uninformative genes (predicted, non-coding, artifacts) from analysis

Installation

# Clone the repository
git clone https://github.com/yourusername/sjanpy.git
cd sjanpy

# Install dependencies
pip install numpy pandas scipy matplotlib seaborn scanpy anndata adjustText statsmodels scikit-learn

Quick Start

import scanpy as sc
from sjanpy import nebulosa, dotplot, deg, embedding

# Load your AnnData object
adata = sc.read_h5ad("your_data.h5ad")

# Nebulosa density plot
nebulosa.nebulosa_density(adata, coord_key="X_umap", gene="CD3D", show=True)

# Complex dot plot with clustering
dotplot.complex_dotplot(adata, genes=marker_genes, groupby="cell_type")

# Differential expression analysis
results = deg.fast_two_group_deg(adata, label_col="condition", lst1=["Disease"], lst2=["Control"])

# High-quality embedding
embedding.fancy_embedding_pro(adata, basis="umap", color="cell_type")

Modules

Module	Description
nebulosa	Weighted 2D KDE for gene expression visualization
deg	Differential expression analysis and volcano plots
dotplot	Complex dot plots with hierarchical clustering and fan layouts
embedding	Publication-ready UMAP/t-SNE visualizations
barplot	Stacked bar plots for cell composition
genecraft	Gene filtering utilities for scRNA-seq

Nebulosa: Solving the Overplotting Problem

Traditional scatter plots can obscure gene expression patterns due to point overlap. Nebulosa uses weighted kernel density estimation to reveal true expression distributions.

Before	After

Dependencies

• Python >= 3.8
• numpy
• pandas
• scipy
• matplotlib
• seaborn
• scanpy
• anndata
• adjustText
• statsmodels
• scikit-learn

License

MIT License

Contributing

Contributions are welcome! Please feel free to submit issues or pull requests.