File formats
Immunology pipelines accept tabular CSV/TSV/JSON inputs where possible. Many endpoints also accept inline row arrays.
Immune composition
Recommended columns:
| Column | Purpose |
|---|---|
cell_type | Immune population label |
group | Optional group, condition, or sample |
Annotation
Recommended columns:
| Column | Purpose |
|---|---|
cell_id | Cell identifier |
broad_cell_type | Broad immune label |
| gene columns | Optional expression values for gene program scoring |
group | Optional grouping |
Repertoire
Supported conventions:
| Convention | Example fields |
|---|---|
| 10x VDJ | barcode, chain, cdr3, v_gene, j_gene, raw_clonotype_id |
| AIRR | locus, junction_aa, v_call, j_call, duplicate_count |
| Bulk TCR/BCR | sample_id, clonotype_id, count |
Communication
Expression rows should contain cell IDs and gene columns. Metadata rows should contain the same cell IDs and a cell type column.
Disease workflows
Useful columns:
cell_idsample_idcell_typeimmune_stateclonotype_idpseudotimex,ycytokine_scoredisease_labeldisease_activity_scoremedicationtimepointvaccination_infection_history
Multi-modal profiling
| Modality | Columns |
|---|---|
| ADT | cell_id plus protein columns |
| ATAC | peak_id, gene, cell_id, accessibility |
| Cytometry events | event_id, sample_id, marker columns such as CD3, CD19, CD14, CD56 |
Binary FCS is feature-gated until a FlowUtils-compatible reader is configured.