Sample data
The current Immunology pipelines include small built-in demo inputs for smoke testing. For real studies, use explicit dataset records and file paths or inline rows.
Minimal composition rows
[
{"cell_type": "CD8 T cell", "group": "case"},
{"cell_type": "CD8 T cell", "group": "case"},
{"cell_type": "B cell", "group": "control"}
]
Minimal repertoire rows
[
{"sample_id": "s1", "barcode": "c1", "chain": "TRB", "cdr3": "CASSA", "v_gene": "TRBV1", "j_gene": "TRBJ1"},
{"sample_id": "s2", "barcode": "c2", "chain": "TRB", "cdr3": "CASSA", "v_gene": "TRBV1", "j_gene": "TRBJ1"}
]
Minimal communication rows
Expression:
[
{"cell_id": "t1", "IFNG": 6, "CXCR3": 2},
{"cell_id": "m1", "CXCL9": 4, "IFNGR1": 5}
]
Metadata:
[
{"cell_id": "t1", "cell_type": "T cell", "condition": "case"},
{"cell_id": "m1", "cell_type": "Myeloid", "condition": "case"}
]
Built-in demos
If a supported analysis is launched without explicit rows or paths, the backend may use built-in demo rows. Demo outputs are useful for route and UI validation but should not be used as study evidence.