Clonal Compass

Single-cell immune-repertoire co-pilot — paired scRNA-seq + TCR-seq. This is an exploratory research analysis, not a diagnostic one.

10,232Cells
20,226Genes
23Clusters
11Cell types
5,300TCR+ cells

1. Overview figures

Leiden clusters on the transcriptome (UMAP).
Leiden clusters on the transcriptome (UMAP).
Marker-based cell-type / T-cell subset labels.
Marker-based cell-type / T-cell subset labels.
TCR clone size (log1p of cells per clonotype).
TCR clone size (log1p of cells per clonotype).
Clonal-expansion category per cell.
Clonal-expansion category per cell.

2. Interpretation of the most clonally-expanded clusters

Each interpretation is tagged with the path that produced it. Claims are hedged and cite the specific metric behind them.

Cluster 6: CD8 T effector/memory

Claude API
Provisional labelCD8 T effector/memory
Cells in cluster409
Top marker genes (rank_genes_groups score)CCL5 (+25.88), IL32 (+25.16), CD8A (+24.24), B2M (+23.93), CTSW (+23.27), KLRK1 (+22.23), CD8B (+21.42), HLA-A (+20.23)
TCR coverage373 / 409 (91%)
Clonal expansion60 cells in clones > 1 (16% of TCR+ cells)
Largest clone sizes here9, 9, 9, 9, 9
Interpretation generated by: Claude API

The provisional label of CD8 T effector/memory is consistent with the marker profile: CD8A (+24.24) and CD8B (+21.42) support a CD8 T-cell identity, while CCL5 (+25.88), IL32 (+25.16), CTSW (+23.27), and KLRK1/NKG2D (+22.23) are suggestive of an effector/cytotoxic-memory program. Note that B2M (+23.93) and HLA-A (+20.23) are broadly expressed antigen-presentation genes and are less specific, so they add only weak support to the effector label. TCR coverage is high at 373/409 cells (91%), so the clonality readout for this cluster is reasonably well-powered. Clonal expansion is modest: only 60 cells (16% of TCR+ cells) sit in clones larger than one, which is consistent with limited, low-level expansion rather than a strongly clonally dominated population. The largest clones are small and evenly sized (several clones of 9 cells), suggestive of a few similarly-sized expansions rather than one outgrowth. No VDJdb epitope match was found for this cluster's expanded clones; this is a normal, valid result and simply means no antigen specificity can be predicted here — it neither confirms nor excludes any target. Overall the data are consistent with a modestly expanded CD8 effector/memory population, but any functional interpretation is exploratory only and should not be read as clinical or diagnostic.

No VDJdb match for this cluster’s expanded clones. Absence of a match is a normal, valid outcome — not a failure.

Cluster 8: MAIT

Claude API
Provisional labelMAIT
Cells in cluster173
Top marker genes (rank_genes_groups score)KLRB1 (+21.73), TRAV1-2 (+19.32), GZMA (+19.09), NKG7 (+19.03), GZMK (+18.99), IL7R (+18.07), CCL5 (+17.88), IL32 (+17.76)
TCR coverage160 / 173 (92%)
Clonal expansion34 cells in clones > 1 (21% of TCR+ cells)
Largest clone sizes here8, 8, 8, 8, 8
Interpretation generated by: Claude API

The provisional MAIT label is consistent with the marker profile: KLRB1 (+21.73, encoding CD161) and TRAV1-2 (+19.32, the canonical MAIT TCR α-chain segment) are both among the top-scoring genes, and this combination is suggestive of a MAIT identity rather than proof of it. The cytotoxic/effector genes GZMA (+19.09), NKG7 (+19.03), GZMK (+18.99), and CCL5 (+17.88), together with IL7R (+18.07) and IL32 (+17.76), are consistent with an effector-memory–like state, though these markers are shared with conventional cytotoxic T cells and are not MAIT-specific. TCR coverage is high at 92% (160/173 cells), so the clonal readout is reasonably well supported. Clonal expansion is modest: 34 cells (21% of TCR+ cells) sit in clones larger than one, with the largest clones each containing 8 cells, which is suggestive of some antigen-driven expansion but is not dominant. No VDJdb epitope match was found for this cluster's expanded clones; this is a normal, valid result and simply means no predicted specificity can be assigned here, not that the cells are non-functional. Any functional reading is exploratory only — this is a research analysis, and no clinical or patient-level conclusion should be drawn. Overall the evidence is consistent with an expanded, effector-like MAIT-enriched cluster, but the MAIT call rests mainly on KLRB1/TRAV1-2 and would benefit from confirmation (e.g., MR1-tetramer staining or fuller invariant TCR analysis).

No VDJdb match for this cluster’s expanded clones. Absence of a match is a normal, valid outcome — not a failure.

Cluster 5: CD4 T memory

Claude API
Provisional labelCD4 T memory
Cells in cluster1,025
Top marker genes (rank_genes_groups score)IL32 (+36.79), AQP3 (+34.81), B2M (+33.06), GSTK1 (+32.09), LTB (+31.81), IL7R (+31.27), CRIP1 (+31.22), ITGB1 (+30.59)
TCR coverage1,007 / 1,025 (98%)
Clonal expansion24 cells in clones > 1 (2% of TCR+ cells)
Largest clone sizes here3, 3, 2, 2, 2
Interpretation generated by: Claude API

The provisional label of CD4 T memory is consistent with the marker profile, notably IL7R (+31.27), a memory/naive T-cell receptor gene, and LTB (+31.81) and AQP3 (+34.81), which are frequently enriched in memory CD4 populations. IL32 (+36.79) and CRIP1 (+31.22) are also elevated, broadly supporting a T-cell activation/memory phenotype, though these markers are not lineage-specific on their own. Note that B2M (+33.06) is a housekeeping-type MHC-I gene and adds little cell-type specificity. TCR coverage is high at 98% (1007/1025 cells), so this is a well-sampled αβ T-cell cluster.

Clonal expansion is minimal: only 24 cells (2% of TCR+ cells) sit in clones larger than one, with the biggest clones being just 3 cells. This low expansion is consistent with a resting or diverse memory pool rather than a clonally driven effector response, but the numbers are too small to draw a firm functional conclusion. No VDJdb epitope matches were found for the expanded clones; this is a normal, valid result and simply means no antigen specificity can be predicted here, not that the cells lack specificity. Overall the evidence is suggestive of a polyclonal CD4 memory T-cell cluster; this is an exploratory research interpretation only and supports no clinical or patient-level claims.

No VDJdb match for this cluster’s expanded clones. Absence of a match is a normal, valid outcome — not a failure.

3. Predicted epitope specificities (VDJdb)

17 TCR+ cell(s) carry a CDR3 that exactly matches a VDJdb entry. These are predicted specificities (database hits), not confirmed reactivity:

Matches are exact CDR3 amino-acid hits against VDJdb and indicate possible/predicted antigen specificity only; they require functional validation. A “no match” result is expected and valid.