scanorama.json

{
  "tool_id": "scanorama_integration",
  "name": "Scanorama Integration",
  "category": "integration",
  "framework": "scanorama",
  "function": "sc.external.pp.scanorama_integrate",
  "status": "optional",
  "valid_after": ["feature_selection"],
  "valid_before": ["neighbor_graph"],
  "paradigms": ["all"],
  "parameters": {
    "batch_key": {
      "type": "string",
      "required": true,
      "guidance": "The obs column containing batch labels."
    },
    "basis": {
      "type": "string",
      "default": "X_pca",
      "guidance": "Input embedding to correct."
    },
    "adjusted_basis": {
      "type": "string",
      "default": "X_scanorama",
      "guidance": "Key for the corrected embedding."
    }
  },
  "outputs": {
    "corrected_embedding": "adata.obsm['X_scanorama']"
  },
  "validation": {
    "check": "Batch mixing should improve. Evaluate with UMAP colored by batch."
  },
  "notes": "Scanorama uses mutual nearest neighbors for alignment. Fast and works well when batches share cell types. Less effective when batches have non-overlapping cell populations.",
  "provenance_captures": ["batch_key", "basis", "adjusted_basis", "n_batches"]
}