---
title: "Analyze Single-Cell Data"
description: "Analyze Single-Cell Data: a genomics recipe chaining scanpy."
tags:
  - recipes
  - genomics
tools:
  - scanpy
generated: true
source: quickActionConfig:single-cell-de-enrichment
---

> **Generated recipe**

> Seeded from a curated quick-action prompt. The prompt below runs the real
> routing — copy it, change the target, and run.

**Best fit:** analyze single-cell data — a genomics goal.

**Try this prompt** — [open in a guest chat](/chat/new?guest=true&prompt=Run%20Scanpy%20clustering%20on%20the%20Allen%20Brain%20Atlas%20mouse%20cortex%20single-cell%20data%20(GEO%3A%20GSE71585)%20to%20identify%20cell%20types%2C%20then%20run%20CellRank%20to%20map%20differentiation%20trajectories%20between%20neuronal%20subtypes%2C%20run%20pathway%20enrichment%20on%20marker%20genes%20with%20Enrichr%2C%20and%20fold%20the%20top%20marker%20protein%20for%20each%20cluster%20with%20ESMFold%20to%20assess%20druggability%20with%20P2Rank%20-%20load%20the%20analysis%20into%20the%20sandbox%20and%20use%20scanpy's%20plotting%20to%20generate%20a%20UMAP%20with%20CellRank%20velocity%20arrows%20and%20a%20side-panel%20of%20P2Rank%20druggability%20scores%20for%20each%20cluster's%20marker%2C%20recommending%20the%20top%20druggable%20neuron-subtype%20marker%20for%20a%20selective%20CNS%20drug%3F).

> Run Scanpy clustering on the Allen Brain Atlas mouse cortex single-cell data (GEO: GSE71585) to identify cell types, then run CellRank to map differentiation trajectories between neuronal subtypes, run pathway enrichment on marker genes with Enrichr, and fold the top marker protein for each cluster with ESMFold to assess druggability with P2Rank - load the analysis into the sandbox and use scanpy's plotting to generate a UMAP with CellRank velocity arrows and a side-panel of P2Rank druggability scores for each cluster's marker, recommending the top druggable neuron-subtype marker for a selective CNS drug?

_A real `single-cell-de-enrichment` quick-action prompt — open it in a guest chat, then swap in your own compound or target._

## The prompt

```text
Run Scanpy clustering on the Allen Brain Atlas mouse cortex single-cell data (GEO: GSE71585) to identify cell types, then run CellRank to map differentiation trajectories between neuronal subtypes, run pathway enrichment on marker genes with Enrichr, and fold the top marker protein for each cluster with ESMFold to assess druggability with P2Rank - load the analysis into the sandbox and use scanpy's plotting to generate a UMAP with CellRank velocity arrows and a side-panel of P2Rank druggability scores for each cluster's marker, recommending the top druggable neuron-subtype marker for a selective CNS drug?
```

## What each tool contributes

- [scanpy](/docs/scanpy) — Differential expression, marker gene ranking per cluster
- **mygene query** — Gene annotation with GO terms, pathways, cross-references
- **enrichr enrichment** — KEGG, Reactome, GO pathway enrichment analysis

## Why these tools pair

Scanpy DE identifies marker genes, MyGene annotates them, Enrichr runs pathway enrichment — full functional characterization pipeline

## Data flow

Every tool operates on the same input: `gene_list`.

## Read next

- [Which tool?](/docs/which-tool) — match a single goal to a tool.
- [Workflows](/docs/workflows) — every workflow and recipe.
- [How the orchestrator picks tools](/docs/how-orchestration-works) — the same chain from a plain-language goal.