SpatialCheckpoint

Spatial heterogeneity profiling of immune checkpoints in spatial transcriptomics data.

SpatialCheckpoint is a bioinformatics pipeline that integrates spatial gene expression profiling, consensus clustering, ensemble ML classification, SHAP interpretability, and clinical survival analysis to characterize immune checkpoint heterogeneity across the tumor microenvironment.

---

Installation

pip install spatialcheckpoint

Requirements: Python ≥ 3.10

---

CLI

# Run the built-in demo (no data files needed)
spatialcheckpoint demo

# Download a registered dataset
spatialcheckpoint download BRCA_visium_10x

# Download all BRCA datasets
spatialcheckpoint download all --cancer-type BRCA

# Preprocess raw Visium output or H5AD
spatialcheckpoint preprocess path/to/spaceranger/  data/processed/
spatialcheckpoint preprocess sample.h5ad           data/processed/

# Run full spatial analysis on a preprocessed sample
spatialcheckpoint analyze sample01

# Discover archetypes from a feature matrix CSV
spatialcheckpoint discover results/sample01/features.csv --k-min 2 --k-max 8

# Train the archetype classifier
spatialcheckpoint classify features.csv archetype_labels.csv --model-dir models/

# Generate publication figures
spatialcheckpoint figures --results-dir results/ --output-dir paper/figures/

---

Usage

Gene Panel

import spatialcheckpoint as scp

# 44 checkpoint genes across 6 functional categories
genes = scp.get_all_checkpoint_genes()

# Genes by category
pd1_pathway  = scp.get_category_genes("co_inhibitory_receptors")
novel        = scp.get_category_genes("novel_checkpoints")
cell_markers = scp.get_immune_cell_markers()    # {cell_type: [genes]}
lr_pairs     = scp.get_ligand_receptor_pairs()  # [{ligand, receptor, alias}]

Data Preprocessing

# From Space Ranger output directory
preprocessor = scp.SpatialDataPreprocessor(spaceranger_out_path="path/to/spaceranger/output")
adata = preprocessor.load_visium()
adata = preprocessor.quality_control(adata, min_genes=200, max_mt_pct=25.0)
adata = preprocessor.normalize(adata)
adata.write_h5ad("data/processed/sample01_preprocessed.h5ad")

# Or from an existing H5AD
preprocessor = scp.SpatialDataPreprocessor(h5_path="existing_data.h5ad")

Spatial Profiling & Feature Extraction

genes = scp.get_all_checkpoint_genes()

# Region-based expression (tumor_core, invasive_margin, stroma, …)
profiler    = scp.SpatialCheckpointProfiler(adata, genes)
region_expr = profiler.expression_by_region()
hotspots    = profiler.checkpoint_hotspot_detection()   # Moran's I per gene

# 80+ spatial features per slide
engineer = scp.SpatialFeatureEngineer(adata, genes)
features = engineer.extract_all_features(sample_id="sample01")

Archetype Discovery

# feature_matrix: DataFrame (n_samples × n_features)
# sample_metadata: DataFrame with 'cancer_type' column, same index
discovery = scp.SpatialArchetypeDiscovery(feature_matrix, sample_metadata)

cc     = discovery.consensus_clustering(k_range=(2, 8), n_iterations=100)
labels = cc["labels"]
char   = discovery.characterize_archetypes(labels)

nmf = discovery.run_nmf(k=cc["optimal_k"])
# nmf["W"]  →  (n_samples, k) soft membership weights
# nmf["H"]  →  (k, n_features) archetype profiles

Classifier Training & SHAP

trainer = scp.ArchetypeModelTrainer(
    feature_matrix=feature_matrix,
    archetype_labels=labels,
    output_dir="models/",
)
results = trainer.run(n_optuna_trials=30)

explainer = scp.ArchetypeExplainer(results["model"], feature_matrix)
shap_df   = explainer.global_feature_importance()

---

Demo

Runs entirely on synthetic data — no Visium files required.

import numpy as np
import pandas as pd
import scanpy as sc
import spatialcheckpoint as scp

print(f"SpatialCheckpoint v{scp.__version__}")

# ── Gene panel ───────────────────────────────────────────────────────────────
genes = scp.get_all_checkpoint_genes()
print(f"Checkpoint panel : {len(genes)} genes")
print(f"PD-1 pathway     : {scp.get_category_genes('co_inhibitory_receptors')}")

lr = scp.get_ligand_receptor_pairs()
print(f"LR pairs ({len(lr)}) e.g. {lr[0]}")

# ── Synthetic Visium slide ───────────────────────────────────────────────────
rng = np.random.default_rng(42)
cp8 = genes[:8]
dummy_genes = [f"GENE{i:04d}" for i in range(92)] + cp8

X = rng.negative_binomial(2, 0.5, size=(200, 100)).astype(float)
adata = sc.AnnData(X=X)
adata.var_names = pd.Index(dummy_genes)

gx, gy = np.meshgrid(np.arange(20), np.arange(10))
coords = np.column_stack([gx.ravel(), gy.ravel()]).astype(float)
coords += rng.uniform(-0.1, 0.1, size=coords.shape)
adata.obsm["spatial"] = coords

region_map = []
for x, y in coords:
    if   x < 5  and y < 5:  region_map.append("tumor_core")
    elif x < 10 and y < 8:  region_map.append("invasive_margin")
    elif x >= 15:            region_map.append("immune_enriched")
    elif y >= 8:             region_map.append("necrotic")
    else:                    region_map.append("stroma")

adata.obs["region_type"] = pd.Categorical(
    region_map,
    categories=["tumor_core","invasive_margin","stroma","immune_enriched","necrotic"]
)

# ── Spatial feature extraction ───────────────────────────────────────────────
engineer = scp.SpatialFeatureEngineer(adata, cp8)
features = engineer.extract_all_features(sample_id="demo")
print(f"\nFeature matrix   : {features.shape[1]} features extracted")

# ── Multi-sample archetype discovery ────────────────────────────────────────
feat_mat = pd.DataFrame(
    rng.standard_normal((30, features.shape[1])),
    index=[f"sample_{i:03d}" for i in range(30)],
    columns=features.columns,
)
meta = pd.DataFrame(
    {"cancer_type": rng.choice(["BRCA","CRC","NSCLC"], 30)},
    index=feat_mat.index,
)

discovery = scp.SpatialArchetypeDiscovery(feat_mat, meta)
cc = discovery.consensus_clustering(k_range=(2, 4), n_iterations=20)
print(f"\nOptimal k        : {cc['optimal_k']}")

char = discovery.characterize_archetypes(cc["labels"])
print("\nArchetype summary:")
print(char[["archetype_name","n_samples"]].to_string())

# ── NMF soft membership ──────────────────────────────────────────────────────
nmf = discovery.run_nmf(k=cc["optimal_k"])
print(f"\nNMF explained variance : {nmf['explained_variance']:.3f}")
print("Membership weights (first 3 samples):")
print(nmf["W"].head(3).round(3).to_string())

---

Gene Panel

Category	Genes (examples)
Co-inhibitory receptors	PDCD1 (PD-1), CTLA4, LAG3, HAVCR2 (TIM-3), TIGIT
Co-inhibitory ligands	CD274 (PD-L1), PDCD1LG2 (PD-L2), LGALS9
Novel checkpoints	VSIR (VISTA), CD276 (B7-H3), VTCN1 (B7-H4)
Innate checkpoints	CD47, SIRPA, LILRB1, LILRB2
Immune enzymes	IDO1, ENTPD1 (CD39), NT5E (CD73), ARG1
Co-stimulatory reference	CD28, ICOS, TNFRSF4 (OX40), TNFRSF9 (4-1BB)

---

Archetypes

Archetype	Spatial signature
Checkpoint-Hot	High checkpoint + high immune + co-localized
Checkpoint-Cold	Low checkpoint + low immune infiltration
Checkpoint-Excluded	Checkpoint at margin, immune at periphery
Checkpoint-Mismatch	Checkpoint and immune spatially separated
Innate-Dominant	CD47/SIRPα axis dominant
Novel-Enriched	VISTA / B7-H3 / B7-H4 enriched

---

Pipeline Architecture

Raw Visium data (Space Ranger dir or H5AD)
  → SpatialDataPreprocessor      QC, normalize → 'counts' / 'log1p' layers
  → SpatialCheckpointProfiler    region expression (tumor_core, invasive_margin,
                                  stroma, immune_enriched, necrotic)
  → SpatialFeatureEngineer       80+ features: co-localization, gradients,
                                  Moran's I, region expression ratios
  → SpatialArchetypeDiscovery    consensus KMeans + delta-area k-selection + NMF
  → ArchetypeModelTrainer        LightGBM + XGBoost + MLP + RF ensemble,
                                  SMOTE, RFECV, Optuna HPO
  → ArchetypeExplainer           SHAP global / per-class feature importance
  → ClinicalAssociationAnalyzer  KM curves, Cox PH, logistic regression (OS/PFS)

---

Output Files

Path	Contents
results/{sample_id}/features.csv	80+ spatial features
results/{sample_id}/region_expression.csv	Region × gene expression
results/{sample_id}/hotspots.csv	Moran's I per gene
results/{sample_id}/colocalization.csv	Ligand-receptor co-occurrence
results/archetypes/archetype_labels.csv	Sample → archetype assignment
results/archetypes/nmf_W.csv, nmf_H.csv	NMF basis / coefficient matrices
models/archetype_classifier.joblib	Serialized ensemble model
paper/figures/	Publication-ready PDF/PNG plots

---

Citation

@article{spatialcheckpoint2025,
  title   = {SpatialCheckpoint: Spatial heterogeneity profiling of immune checkpoints
             in spatial transcriptomics},
  author  = {},
  journal = {},
  year    = {2025},
}

---

License

MIT