Strand AI predicts protein markers directly from an H&E whole-slide image using POSTMAN, our spatial-proteomics imputation model. Each marker listed below has a model checkpoint trained against paired multiplex ground truth. To request a marker not on this list, email support@strandai.com. We add markers as ground-truth training data becomes available.Documentation Index
Fetch the complete documentation index at: https://docs.strandai.com/llms.txt
Use this file to discover all available pages before exploring further.
Per-marker model cards with calibration plots, per-tissue accuracy, and
held-out benchmarks are coming soon. The table below is a snapshot of
what’s predictable today.
Marker panel
All 52 markers below are served bypostman-v1. Grouped by biology; expand any
group to see the full list.
Immune lineage (18)
Immune lineage (18)
| Marker | Predicts |
|---|---|
| CD3e | Pan-T cell marker. |
| CD4 | Helper T cells. |
| CD8 | Cytotoxic T cells. |
| CD45 | Pan-leukocyte. |
| CD45RA | Naive T cells (PTPRC isoform). |
| CD45RO | Memory T cells (PTPRC isoform). |
| CD20 | Mature B cells. |
| CD21 | Mature B cells and follicular dendritic cells. |
| CD79 | B cell receptor signaling. |
| CD11b | Myeloid / neutrophils / monocytes. |
| CD11c | Dendritic cell / myeloid lineage marker. |
| CD14 | Monocytes and tissue macrophages. |
| CD68 | Pan-macrophage. |
| CD163 | M2-polarised macrophages. |
| CD141 | Endothelial and BDCA-3 dendritic cell subset. |
| CD66 | Granulocytes and select epithelial cells (CEACAM family). |
| MPO | Neutrophil granule enzyme. |
| FoxP3 | Regulatory T cells. |
Activation, checkpoint & function (15)
Activation, checkpoint & function (15)
| Marker | Predicts |
|---|---|
| PD1 | T-cell exhaustion / checkpoint. |
| PDL1 | Tumor and immune PD-L1 expression. |
| LAG3 | T cell exhaustion / checkpoint. |
| VISTA | T cell suppression / checkpoint. |
| ICOS | T cell costimulation. |
| CD38 | B, T, and plasma cell activation. |
| CD39 | T-regulatory / endothelial ectonucleotidase. |
| CD40 | B cell activation and antigen-presenting cells. |
| CD44 | Hyaluronan receptor; broad immune and tumor expression. |
| IDO1 | Immunosuppressive enzyme. |
| GranzymeB | Cytotoxic effector function. |
| IFNg | Interferon-γ; activated T and NK cells. |
| HLA-DR | MHC class II; antigen-presenting cells. |
| HLA-ABC | Classical MHC class I. |
| HLA-E | Non-classical MHC class I; NK cell inhibition. |
Nuclear & proliferation (3)
Nuclear & proliferation (3)
| Marker | Predicts |
|---|---|
| DAPI | Nuclear counterstain (sanity-check channel). |
| Ki67 | Proliferation marker. |
| PCNA | Proliferation marker (DNA replication). |
Epithelial & tumor (6)
Epithelial & tumor (6)
| Marker | Predicts |
|---|---|
| PanCK | Pan-cytokeratin; epithelial / tumor compartment. |
| EpCAM | Pan-epithelial adhesion marker. |
| ECad | E-cadherin; epithelial cell-cell adhesion. |
| Keratin8/18 | Simple epithelial cytokeratins. |
| TP63 | Squamous and basal epithelial cells. |
| GATA3 | TH2 / luminal epithelial transcription factor. |
Stromal, vascular & structural (7)
Stromal, vascular & structural (7)
| Marker | Predicts |
|---|---|
| aSMA | α-smooth muscle actin; stromal myofibroblasts and vasculature. |
| Vimentin | Mesenchymal marker; stromal and EMT-state cells. |
| Caveolin1 | Caveolae scaffold; stromal and tumor compartments. |
| CD31 | PECAM-1; endothelium and microvasculature. |
| CD34 | Endothelial and hematopoietic stem / progenitor cells. |
| CollagenIV | Basement membrane. |
| Podoplanin | Lymphatic endothelium and cancer-associated stroma. |
Other (3)
Other (3)
| Marker | Predicts |
|---|---|
| BCL2 | Apoptosis regulator. |
| Gal3 | Galectin-3; immune and stromal modulation. |
| PGP9.5 | Pan-neuronal / nerve fibers. |
Reading the predictions
Predictions are returned as a multi-channel OME-Zarr aligned to the slide’s pixel grid. Each requested marker becomes a separate channel; the SDK helpers convert it toAnnData (Python) or SpatialExperiment (R) so
you can treat it like real multiplex data.
See the Quickstart for an end-to-end example and the
Python / R SDK reference for the conversion
helpers.
Accuracy caveats
- Predictions are model outputs, not ground truth. Use them as a hypothesis-generation surface, particularly on cohorts that look out-of-distribution to the training data.
- Per-marker confidence varies. Lineage and structural markers (e.g. PanCK, CD31, aSMA) generally calibrate better than functional / activation markers (e.g. GranzymeB, PD1).
- We do not yet publish per-tissue or per-organ accuracy numbers. Those ship with the per-marker model cards.