Generate single-cell RNA-seq analysis code templates for Seurat and Scanpy, supporting QC, clustering, visualization, and downstream analysis. Trigger when users need scRNA-seq analysis pipelines, preprocessing workflows, or batch correction code.
Generate comprehensive single-cell RNA-seq analysis code templates for Seurat (R) and Scanpy (Python). This skill provides ready-to-use code frameworks for preprocessing, quality control, normalization, clustering, marker identification, visualization, and advanced analyses like batch correction and trajectory inference.
Technical Difficulty: High
python scripts/main.py --tool seurat --output seurat_analysis.R --species human
python scripts/main.py --tool scanpy --output scanpy_analysis.py --species mouse
python scripts/main.py --tool both --output scrna_pipeline --species human --batch-correction harmony --trajectory true
| Parameter | Type | Required | Description |
|---|---|---|---|
| --tool | string | Yes | Analysis tool: seurat, scanpy, or both |
| --output | string | Yes | Output file or directory path |
| --species | string | No | Species: human or mouse (default: human) |
| --batch-correction | string | No | Method: harmony, rpca, cca, scanorama, scvi |
| --trajectory | bool | No | Include trajectory analysis (default: false) |
| --cell-communication | bool | No | Include cell-cell communication (default: false) |
| --de-analysis | bool | No | Include differential expression (default: false) |
| --spatial | bool | No | Include spatial transcriptomics (default: false) |
output/
├── seurat/
│ ├── 01_load_and_qc.R
│ ├── 02_normalize_integrate.R
│ ├── 03_cluster_annotate.R
│ ├── 04_visualize.R
│ └── 05_de_analysis.R (if --de-analysis)
├── scanpy/
│ ├── 01_load_qc.py
│ ├── 02_normalize_integrate.py
│ ├── 03_cluster_annotate.py
│ ├── 04_visualize.py
│ └── 05_trajectory.py (if --trajectory)
└── README.md
| Metric | Human | Mouse |
|---|---|---|
| min_genes | 200 | 200 |
| max_genes | 25000 | 25000 |
| min_cells | 3 | 3 |
| max_mt_percent | 20% | 20% |
| doublet_threshold | Auto | Auto |
| Scenario | Seurat | Scanpy |
|---|---|---|
| Small batches (<5) | Harmony | Harmony |
| Large batches | RPCA | Scanorama |
| Complex variation | CCA | scVI |
# Load data
seurat_obj <- CreateSeuratObject(counts = raw_data, project = "Sample")
# QC
seurat_obj[["percent.mt"]] <- PercentageFeatureSet(seurat_obj, pattern = "^MT-")
seurat_obj <- subset(seurat_obj, subset = nFeature_RNA > 200 & percent.mt < 20)
# Normalize
seurat_obj <- NormalizeData(seurat_obj)
seurat_obj <- FindVariableFeatures(seurat_obj, selection.method = "vst", nfeatures = 2000)
# Scale and PCA
seurat_obj <- ScaleData(seurat_obj)
seurat_obj <- RunPCA(seurat_obj, features = VariableFeatures(object = seurat_obj))
# Cluster
seurat_obj <- FindNeighbors(seurat_obj, dims = 1:30)
seurat_obj <- FindClusters(seurat_obj, resolution = 1.0)
seurat_obj <- RunUMAP(seurat_obj, dims = 1:30)
# Visualize
DimPlot(seurat_obj, reduction = "umap", label = TRUE)
FeaturePlot(seurat_obj, features = c("CD3E", "CD14", "CD79A"))
import scanpy as sc
# Load data
adata = sc.read_10x_mtx("filtered_gene_bc_matrices/")
# QC
sc.pp.filter_cells(adata, min_genes=200)
sc.pp.filter_genes(adata, min_cells=3)
adata.var['mt'] = adata.var_names.str.startswith('MT-')
sc.pp.calculate_qc_metrics(adata, qc_vars=['mt'], percent_top=None, inplace=True)
adata = adata[adata.obs.pct_counts_mt < 20, :]
# Normalize
sc.pp.normalize_total(adata, target_sum=1e4)
sc.pp.log1p(adata)
sc.pp.highly_variable_genes(adata, n_top_genes=2000)
# PCA and UMAP
sc.pp.scale(adata)
sc.tl.pca(adata, svd_solver='arpack')
sc.pp.neighbors(adata, n_neighbors=15, n_pcs=30)
sc.tl.umap(adata)
sc.tl.leiden(adata, resolution=1.0)
# Visualize
sc.pl.umap(adata, color=['leiden', 'total_counts'])
sc.pl.dotplot(adata, var_names=['CD3E', 'CD14', 'CD79A'], groupby='leiden')
references/seurat_template.R - Complete Seurat analysis templatereferences/scanpy_template.py - Complete Scanpy analysis templatereferences/batch_correction_guide.md - Batch correction comparisonrequirements.txt - Python dependenciesinstall.packages(c("Seurat", "SeuratObject", "tidyverse", "patchwork"))
# Optional