Squidpy Pack 2 — Deconvolution & Spatial Pattern Detection

**2 in-depth Jupyter notebooks covering Visium spot deconvolution and spatially variable gene detection — with all outputs already included.** Both notebooks run on the Squidpy built-in mouse H&E Visium dataset. No data downloads required. ### What's inside **01 — cell2location: Deconvolving Visium Spots to Single-Cell Resolution** (~45 min CPU / ~10 min GPU) - Why Visium spots need deconvolution and how NB regression works - Preparing a matched scRNA-seq reference - Training the reference and spatial mapping models - Mapping 5 cell types (cancer, T cells, macrophages, fibroblasts, endothelial) - Identifying tissue zones by cell type composition - Spatial neighborhood enrichment between zones **02 — SpatialDE2: Finding Spatially Variable Genes Without Clustering** (~10 min CPU) - Moran's I for fast SVG screening - SpatialDE2 pattern classification (hotspot / gradient / SE) - Linking SVGs back to cell type markers - Spatial co-expression module detection - Speed vs. accuracy tradeoffs between methods ### Who this is for - Researchers who need cell-type resolution from Visium spot data - Anyone moving beyond basic clustering to spatially variable gene detection - Squidpy users ready for the next layer past Pack 1 Foundations ### Requirements ``` pip install cell2location[tutorials] SpatialDE2 squidpy scanpy matplotlib ```

Found on Devadex — the discovery index for independent software the big search engines bury. More from gumroad.

Report this listing