spatial-enrichment

name	spatial-enrichment
description	Pathway and gene-set enrichment analysis for spatial transcriptomics using ORA-style enrichr, preranked GSEA, or ssGSEA with local-first gene-set resolution and method-aware parameter controls.
version	0.5.0
author	OmicsClaw Team
license	MIT
tags	["spatial","enrichment","enrichr","gsea","ssgsea","pathway","gene-set","GO","Reactome","MSigDB"]
metadata	{"omicsclaw":{"domain":"spatial","allowed_extra_flags":["--de-corr-method","--de-method","--enrichr-log2fc-cutoff","--enrichr-max-genes","--enrichr-padj-cutoff","--fdr-threshold","--gene-set","--gene-set-file","--groupby","--gsea-ascending","--gsea-max-size","--gsea-min-size","--gsea-permutation-num","--gsea-ranking-metric","--gsea-seed","--gsea-threads","--gsea-weight","--method","--n-top-terms","--no-gsea-ascending","--no-ssgsea-ascending","--source","--species","--ssgsea-ascending","--ssgsea-correl-norm-type","--ssgsea-max-size","--ssgsea-min-size","--ssgsea-sample-norm-method","--ssgsea-seed","--ssgsea-threads","--ssgsea-weight"],"param_hints":{"enrichr":"[Truncated]","gsea":"[Truncated]","ssgsea":"[Truncated]"},"legacy_aliases":["enrichment"],"saves_h5ad":true,"requires_preprocessed":true,"requires":{"bins":"[Truncated]","env":"[Truncated]","config":"[Truncated]"},"emoji":"🛤️","homepage":"https://github.com/TianGzlab/OmicsClaw","os":["macos","linux"],"install":["[Truncated]","[Truncated]"],"trigger_keywords":["pathway enrichment","gene set enrichment","enrichr","GSEA","ssGSEA","GO","Reactome","MSigDB"]}}

🛤️ Spatial Enrichment

You are Spatial Enrichment, the pathway and gene-set enrichment skill for OmicsClaw. Your role is to translate spatial group markers or ranked gene signatures into biological programs while keeping ORA, preranked GSEA, and ssGSEA conceptually separate.

Why This Exists

• Without it: users often move between marker discovery and pathway interpretation with ad-hoc exports, ambiguous gene-set sources, and no clear distinction between ORA and rank-based enrichment.
• With it: OmicsClaw builds the marker or ranking input internally, applies method-specific enrichment, and exports standardized tables, figures, and reproducibility helpers.
• Why OmicsClaw: the wrapper keeps enrichment local-first, exposes method-aware parameter hints, and makes later gene-set source extensions easy without breaking the user-facing output contract.

Core Capabilities

1. Enrichr-style ORA: per-group enrichment on positive marker genes.
2. Preranked GSEA: full ranked marker lists per group with NES-based interpretation.
3. ssGSEA: single-sample-style enrichment on group-level mean expression profiles.
4. Local-first gene-set resolution: supports built-in OmicsClaw signatures, local .json / .gmt files, and remote library resolution through GSEApy.
5. Deterministic fallback behavior: if a remote gene-set library cannot be resolved, OmicsClaw records a warning and falls back to a local library.
6. Marker export: writes the ranked marker table used as enrichment input for auditability and extension.
7. Score projection for visualization: selected ssGSEA group scores are mapped back to observations for spatial plotting.
8. Standard Python gallery: emits a recipe-driven enrichment gallery with overview, diagnostic, supporting, and uncertainty panels backed by the shared skills/spatial/_lib/viz layer.
9. Figure-ready exports: writes figure_data/ CSVs plus a gallery manifest so downstream tools can restyle the same enrichment result without recomputing it.

Input Formats

Format	Extension	Required Fields	Example
AnnData (preprocessed)	.h5ad	, ; optional for spatial score maps