CyteType returns detailed annotations per cluster and stores them on your AnnData object.
The get_results() method is the recommended way to retrieve CyteType results. It provides intelligent result fetching with automatic fallback to API retrieval:
# Helper
results = annotator.get_results()
import json
results = json.loads(adata.uns["cytetype_results"]["result"])- annotation: primary cell type
- granularAnnotation: finer phenotype
- cellState: state/activation/malignancy
- ontologyTerm: Cell Ontology term (e.g., CL_0000127)
- confidence: model-assessed confidence
- supportingMarkers / conflictingMarkers
- missingExpression / unexpectedExpression
- corroboratingPapers: literature support
Example loop:
for ann in results["annotations"]:
print(ann["clusterId"], ann["annotation"], ann["ontologyTerm"], ann["confidence"]) For end-to-end examples and reports, see Examples.