scverse/squidpy

What it solves

Squidpy provides a scalable framework for analyzing and visualizing spatial molecular data. It addresses the challenge of processing high-resolution tissue microscopy images and spatial omics assays (such as Visium, Slide-seq, and Xenium) to understand how cells and genes are organized within a tissue section.

How it works

Building on the scanpy and anndata ecosystems, Squidpy provides streamlined APIs to extract features from images and compute spatial statistics. It allows users to build spatial neighbor graphs and analyze the co-occurrence and enrichment of different cell types and genes across a tissue.

Who it’s for

It is designed for researchers and scientists working with spatial omics and tissue microscopy to analyze the spatial organization of molecular data.

Highlights

• Spatial Neighbor Graphs: Build and analyze graphs from various spatial omics assays.
• Spatial Statistics: Compute neighborhood enrichment, co-occurrence, and Moran's I for cell types and genes.
• Image Integration: Efficiently store, featurize, and visualize high-resolution microscopy images using scikit-image.
• Interactive Exploration: Integration with napari-spatialdata for interactive dataset exploration.