sc-enrichment

name	sc-enrichment
description	Statistical enrichment analysis for single-cell RNA-seq using ORA or preranked GSEA on marker or differential-expression rankings. This skill is for GO/KEGG/Reactome/Hallmark term significance, not per-cell pathway activity scoring.
version	0.1.0
author	OmicsClaw
license	MIT
tags	["singlecell","enrichment","ORA","GSEA","GO","KEGG","Reactome","Hallmark"]
metadata	{"omicsclaw":{"domain":"singlecell","allowed_extra_flags":["--gene-set-db","--gene-set-from-markers","--gene-sets","--engine","--groupby","--gsea-max-size","--gsea-min-size","--gsea-permutation-num","--gsea-ranking-metric","--gsea-seed","--gsea-weight","--marker-group","--marker-top-n","--method","--ora-log2fc-cutoff","--ora-max-genes","--ora-padj-cutoff","--ranking-method","--species","--top-terms","--r-enhanced"],"param_hints":{"ora":"[Truncated]","gsea":"[Truncated]"},"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":["sc enrichment","single-cell enrichment","GO enrichment","KEGG enrichment","GSEA","ORA","pathway enrichment"]}}

Single-Cell Statistical Enrichment

Why This Exists

• Without it: users jump from marker genes to pathway interpretation with ad-hoc exports and no clear distinction between statistical enrichment and per-cell pathway scoring.
• With it: OmicsClaw can reuse sc-markers / sc-de outputs or auto-rank a clustered h5ad, then run ORA or preranked GSEA with standardized outputs.
• Why OmicsClaw: it keeps the workflow local-first, makes upstream/downstream guidance explicit, and preserves the singlecell output contract.

Core Capabilities

1. ORA on positive marker / DEG lists.
2. Preranked GSEA on full group-specific rankings.
3. Flexible upstream intake:
   • output directory from sc-markers
   • output directory from sc-de
   • processed .h5ad with automatic cluster-vs-rest ranking
4. Marker-derived custom gene sets:
   • convert one or more sc-markers groups into gene sets directly
5. Local-first gene-set sources:
   • local .gmt / .json
   • built-in library keys via --gene-set-db
6. Standard outputs:
   • processed.h5ad
   • figures, tables, figure-ready CSV exports
   • report and reproducibility helpers

Scope Boundary

This skill performs statistical enrichment.

• Use sc-enrichment when you want GO / KEGG / Reactome / Hallmark terms that are statistically over-represented or enriched.
• Use sc-pathway-scoring when you want a pathway/signature score for each individual cell.

In plain language:

• sc-enrichment answers: “which biological terms are significantly enriched?”
• sc-pathway-scoring answers: “how active is this gene signature in each cell?”

Input Formats

Format	Supported?	Notes