Identify spatially variable genes in spatial transcriptomics data with SpatialDE.
npx skills add https://github.com/chenyhvvvv/STAT-agent --skill svg-spatialde
Please help me install this Agent Skill. Skill: svg-spatialde Source: https://github.com/chenyhvvvv/STAT-agent/tree/main/stat_agent/skills/svg-SpatialDE Command: npx skills add https://github.com/chenyhvvvv/STAT-agent --skill svg-spatialde
Spatially resolved gene expression data often contain genes whose expression patterns vary across tissue space; SVGs reveal spatial organization and tissue architecture.
Run SpatialDE on your spatial transcriptomics dataset to identify SVGs and view per-gene results in adata.uns/spatialde_results.
High-intent search queries and answers about installing and using this skill.
You identify spatially variable genes by testing each gene for spatial expression patterns using Gaussian process regression. This method returns per-gene statistics, including p-values, FDR q-values, and spatial length scales, to reveal tissue architecture.
Spatial length scale is a per-gene statistic reported during spatially variable gene identification. It measures the spatial range over which a gene's expression varies across tissue space, helping characterize the spatial organization of the tissue.
Yes, you can find SVGs in cell-level data with appropriate preprocessing. While primarily applicable to spot-level data like Visium, cell-level spatial transcriptomics requires specific preprocessing steps to yield accurate per-gene statistics.
Yes, NaiveDE preprocessing is required before running SpatialDE. This preparatory step normalizes the spatial transcriptomics data so the Gaussian process regression can accurately compute per-gene p-values, q-values, and length scales.
SpatialDE gene expression variance results are stored directly in the AnnData object. Specifically, the per-gene statistics including p-values, FDR q-values, and spatial length scales are saved in the adata.uns dictionary under the spatialde_results key.
TianGzlabDetect spatially variable genes in AnnData objects using Moran's I, SpatialDE, SPARK-X, and FlashS.
chenyhvvvvIdentify differentially expressed genes across defined groups in AnnData objects.
ya-wayOverlay gene expression and annotations on tissue sections using Squidpy and Scanpy.
TianGzlabInfer copy number variations from spatial transcriptomics data using inferCNVpy or Numbat.
