K-Dense-AI/scientific-agent-skills / skills/scanpy
scanpy
VerifiedPerforms Scanpy single-cell RNA-seq QC, normalization, HVG selection, PCA/UMAP/t-SNE, clustering, exploratory marker ranking, pseudobulk preparation, visualization, and Seurat or SingleCellExperiment RDS conversion to h5ad. Applies to established exploratory scRNA-seq workflows with explicit count and expression provenance; complementary skills cover scvi-tools models and AnnData format details.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 19.8 KB |
| assets/analysis_template.py | script | 10.0 KB |
| assets/celltype_mapping.json | config | 184 B |
| assets/gene_signatures.json | config | 355 B |
| assets/pipeline_config.json | config | 380 B |
| references/analysis_workflow.md | markdown | 7.6 KB |
| references/api_reference.md | markdown | 9.1 KB |
| references/plotting_guide.md | markdown | 10.5 KB |
| references/r_interop.md | markdown | 11.8 KB |
| references/standard_workflow.md | markdown | 7.0 KB |
| references/upstream-review.md | markdown | 8.3 KB |
| scripts/_common.py | script | 10.0 KB |
| scripts/annotate.py | script | 3.7 KB |
| scripts/batch_correct.py | script | 3.1 KB |
| scripts/cluster.py | script | 2.5 KB |
| scripts/convert.py | script | 1.5 KB |
| scripts/find_markers.py | script | 4.3 KB |
| scripts/inspect_data.py | script | 3.1 KB |
| scripts/plot.py | script | 3.6 KB |
| scripts/preprocess.py | script | 3.5 KB |
| scripts/pseudobulk.py | script | 4.6 KB |
| scripts/qc_analysis.pyexecutable | script | 5.1 KB |
| scripts/reduce_dimensions.py | script | 2.8 KB |
| scripts/run_pipeline.py | script | 9.4 KB |
| scripts/score_genes.py | script | 3.9 KB |
| scripts/subset.py | script | 2.6 KB |
assets/analysis_template.py
View on GitHub#!/usr/bin/env python3
"""
Complete Single-Cell Analysis Template
This template provides a complete workflow for single-cell RNA-seq analysis
using scanpy, from data loading through clustering and cell type annotation.
Illustrative template targeting Scanpy 1.12.4. Customize thresholds, metadata,
and annotation; this file is not an automatically validated biological analysis.
"""
import scanpy as sc
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
# ============================================================================
# CONFIGURATION
# ============================================================================
# File paths
INPUT_FILE = 'data/raw_counts.h5ad' # Change to your input file
OUTPUT_DIR = 'results/'
FIGURES_DIR = 'figures/'
# QC parameters
MIN_GENES = 200 # Minimum genes per cell
MIN_CELLS = 3 # Minimum cells per gene
MT_THRESHOLD = 5 # Maximum mitochondrial percentage
# Analysis parameters
N_TOP_GENES = 2000 # Number of highly variable genes
N_PCS = 40 # Number of principal components
N_NEIGHBORS = 10 # Number of neighbors for graph
LEIDEN_RESOLUTION = 0.5 # Clustering resolution
# Scanpy settings
sc.settings.verbosity = 3
sc.set_figure_params(dpi=80, facecolor='white')
sc.settings.figdir = FIGURES_DIR
sc.settings.autosave = True
# ============================================================================
# 1. LOAD DATA
# ============================================================================
print("=" * 80)
print("LOADING DATA")
print("=" * 80)
# Load data (adjust based on your file format)
adata = sc.read_h5ad(INPUT_FILE)
# adata = sc.read_10x_mtx('data/filtered_gene_bc_matrices/') # For 10X data
# adata = sc.read_csv('data/counts.csv') # For CSV data
print(f"Loaded: {adata.n_obs} cells x {adata.n_vars} genes")
# ============================================================================
# 2. QUALITY CONTROL
# ============================================================================
print("\n" + "=" * 80)
print("QUALITY CONTROL")
print("=" * 80)
# Identify mitochondrial genes
adata.var['mt'] = adata.var_names.str.startswith('MT-')
# Calculate QC metrics
sc.pp.calculate_qc_metrics(adata, qc_vars=['mt'], percent_top=None,
log1p=False, inplace=True)
# Visualize QC metrics before filtering
sc.pl.violin(adata, ['n_genes_by_counts', 'total_counts', 'pct_counts_mt'],
jitter=0.4, multi_panel=True, save='_qc_before_filtering')
sc.pl.scatter(adata, x='total_counts', y='pct_counts_mt', save='_qc_mt')
sc.pl.scatter(adata, x='total_counts', y='n_genes_by_counts', save='_qc_genes')
# Filter cells and genes
print(f"\nBefore filtering: {adata.n_obs} cells, {adata.n_vars} genes")
sc.pp.filter_cells(adata, min_genes=MIN_GENES)
sc.pp.filter_genes(adata, min_cells=MIN_CELLS)
adata = adata[adata.obs.pct_counts_mt < MT_THRESHOLD, :].copy()
print(f"After filtering: {adata.n_obs} cells, {adata.n_vars} genes")
# Optional: doublet detection (uncomment; run before normalization)
# sc.pp.scrublet(adata)
# adata = adata[~adata.obs['predicted_doublet'], :].copy()
# print(f"After doublet removal: {adata.n_obs} cells")
# ============================================================================
# 3. NORMALIZATION
# ============================================================================
print("\n" + "=" * 80)
print("NORMALIZATION")
print("=" * 80)
# Identified raw counts: preserve before any normalization
adata.layers["counts"] = adata.X.copy()
# Normalize to 10,000 counts per cell
sc.pp.normalize_total(adata, target_sum=1e4)
# Log-transform
sc.pp.log1p(adata)
# Store normalized data
adata.raw = adata.copy()
# ============================================================================
# 4. FEATURE SELECTION
# ============================================================================
print("\n" + "=" * 80)
print("FEATURE SELECTION")
print("=" * 80)
# Identify highly variable genes
sc.pp.highly_variable_genes(adata, n_top_genes=N_TOP_GENES)
# Visualize
sc.pl.highly_variable_genes(adata, save='_hvg')
print(f"Selected {sum(adata.var.highly_variable)} highly variable genes")
# Keep all genes/counts in the exported object; PCA automatically uses the HVG mask.
# ============================================================================
# 5. SCALING AND REGRESSION
# ============================================================================
print("\n" + "=" * 80)
print("SCALING AND REGRESSION")
print("=" * 80)
# Regress out unwanted sources of variation
# sc.pp.regress_out(adata, ['total_counts', 'pct_counts_mt'])
# Scale data
# sc.pp.scale(adata, max_value=10) # Optional; may densify sparse data
# ============================================================================
# 6. DIMENSIONALITY REDUCTION
# ============================================================================
print("\n" + "=" * 80)
print("DIMENSIONALITY REDUCTION")
print("=" * 80)
# PCA
n_pcs = min(N_PCS, adata.n_obs - 1, int(adata.var.highly_variable.sum()) - 1)
sc.pp.pca(adata, n_comps=n_pcs, svd_solver='arpack')
sc.pl.pca_variance_ratio(adata, log=True, save='_pca_variance')
# Compute neighborhood graph
sc.pp.neighbors(adata, n_neighbors=N_NEIGHBORS, n_pcs=n_pcs, use_rep="X_pca")
# UMAP
sc.tl.umap(adata)
# ============================================================================
# 7. CLUSTERING
# ============================================================================
print("\n" + "=" * 80)
print("CLUSTERING")
print("=" * 80)
# Leiden clustering
sc.tl.leiden(adata, resolution=LEIDEN_RESOLUTION, flavor="igraph", directed=False, n_iterations=2)
# Visualize
sc.pl.umap(adata, color='leiden', legend_loc='on data', save='_leiden')
print(f"Identified {len(adata.obs['leiden'].unique())} clusters")
# ============================================================================
# 8. MARKER GENE IDENTIFICATION
# ============================================================================
print("\n" + "=" * 80)
print("MARKER GENE IDENTIFICATION")
print("=" * 80)
# Find marker genes (exploratory — pseudobulk + pydeseq2 for rigorous DE)
sc.tl.rank_genes_groups(adata, 'leiden', method='wilcoxon', use_raw=True,
reference='rest', corr_method='benjamini-hochberg', tie_correct=True)
# Visualize top markers
sc.pl.rank_genes_groups(adata, n_genes=25, sharey=False, save='_markers')
sc.pl.rank_genes_groups_heatmap(adata, n_genes=10, save='_markers_heatmap')
sc.pl.rank_genes_groups_dotplot(adata, n_genes=5, save='_markers_dotplot')
# Get top markers for each cluster
for cluster in adata.obs['leiden'].unique():
print(f"\nCluster {cluster} top markers:")
markers = sc.get.rank_genes_groups_df(adata, group=cluster).head(10)
print(markers[['names', 'scores', 'pvals_adj']].to_string(index=False))
# ============================================================================
# 9. CELL TYPE ANNOTATION (CUSTOMIZE THIS SECTION)
# ============================================================================
print("\n" + "=" * 80)
print("CELL TYPE ANNOTATION")
print("=" * 80)
# Example marker genes for common cell types (customize for your data)
marker_genes = {
'T cells': ['CD3D', 'CD3E', 'CD3G'],
'B cells': ['MS4A1', 'CD79A', 'CD79B'],
'Monocytes': ['CD14', 'LYZ', 'S100A8'],
'NK cells': ['NKG7', 'GNLY', 'KLRD1'],
'Dendritic cells': ['FCER1A', 'CST3'],
}
# Visualize marker genes
for cell_type, genes in marker_genes.items():
available_genes = [g for g in genes if g in adata.raw.var_names]
if available_genes:
sc.pl.umap(adata, color=available_genes, use_raw=True,
save=f'_{cell_type.replace(" ", "_")}')
# Manual annotation based on marker expression (customize this mapping)
cluster_to_celltype = {} # Fill only after reviewing markers, provenance and uncertainty.
# Apply annotations
adata.obs['cell_type'] = adata.obs['leiden'].astype(str).map(cluster_to_celltype)
adata.obs['cell_type'] = adata.obs['cell_type'].fillna('Unknown').astype('category')
# Visualize annotated cell types
sc.pl.umap(adata, color='cell_type', legend_loc='on data', save='_celltypes')
# ============================================================================
# 10. ADDITIONAL ANALYSES (OPTIONAL)
# ============================================================================
print("\n" + "=" * 80)
print("ADDITIONAL ANALYSES")
print("=" * 80)
# PAGA trajectory analysis (optional)
# sc.tl.paga(adata, groups='leiden')
# sc.pl.paga(adata, color='leiden', save='_paga')
# Gene set scoring (optional)
# example_gene_set = ['CD3D', 'CD3E', 'CD3G']
# sc.tl.score_genes(adata, example_gene_set, score_name='T_cell_score')
# sc.pl.umap(adata, color='T_cell_score', save='_gene_set_score')
# ============================================================================
# 11. SAVE RESULTS
# ============================================================================
print("\n" + "=" * 80)
print("SAVING RESULTS")
print("=" * 80)
import os
os.makedirs(OUTPUT_DIR, exist_ok=True)
# Save processed AnnData object
adata.write(f'{OUTPUT_DIR}/processed_data.h5ad')
print(f"Saved processed data to {OUTPUT_DIR}/processed_data.h5ad")
# Export metadata
adata.obs.to_csv(f'{OUTPUT_DIR}/cell_metadata.csv')
adata.var.to_csv(f'{OUTPUT_DIR}/gene_metadata.csv')
print(f"Saved metadata to {OUTPUT_DIR}/")
# Export marker genes
for cluster in adata.obs['leiden'].unique():
markers = sc.get.rank_genes_groups_df(adata, group=cluster)
markers.to_csv(f'{OUTPUT_DIR}/markers_cluster_{cluster}.csv', index=False)
print(f"Saved marker genes to {OUTPUT_DIR}/")
# ============================================================================
# 12. SUMMARY
# ============================================================================
print("\n" + "=" * 80)
print("ANALYSIS SUMMARY")
print("=" * 80)
print(f"\nFinal dataset:")
print(f" Cells: {adata.n_obs}")
print(f" Genes: {adata.n_vars}")
print(f" Clusters: {len(adata.obs['leiden'].unique())}")
print(f"\nCell type distribution:")
print(adata.obs['cell_type'].value_counts())
print("\n" + "=" * 80)
print("ANALYSIS COMPLETE")
print("=" * 80)
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |