Introducing POSTMAN: Spatial Proteomics from H&E
POSTMAN is our state-of-the-art AI model that predicts spatial protein expression from routine H&E pathology slides. We're opening a design partner program for early access.
The protein signal is in the slide
H&E staining is cheap, universal, and already collected for virtually every patient who's had a biopsy. But it's traditionally limited to morphology. The molecular detail stays hidden.
Spatial proteomics technologies like CODEX, MIBI, and multiplex IHC show where specific proteins are expressed across a tissue section. They're useful, but slow, expensive, and hard to scale. Most cohorts will never get spatial proteomics data.
We've found that the spatial protein signal is already encoded in the H&E. Tissue morphology (the shape and arrangement of cells, the structure of the microenvironment, spatial patterns of infiltration) reflects underlying molecular state. POSTMAN learns these relationships from paired H&E and spatial proteomics datasets, then predicts spatially-resolved protein expression from standard pathology images.
What this enables
No additional staining, equipment, or tissue. Just the slides you already have.
- Profile cohorts retrospectively. Get spatial protein data from archived slides that were never intended for proteomics.
- Triage before expensive assays. Identify the most promising samples for follow-up with ground-truth spatial proteomics.
- Surface biomarkers you never measured. Find predictive protein signatures across cohorts where spatial proteomics wasn't feasible.
Why this matters for drug development
Relevant biology is encoded in tissue that's already been collected and sitting in archives. With predicted spatial protein data across your cohort, you can stratify patients using spatial phenotypes, identify companion diagnostics from routine pathology, and power studies that were previously limited by missing modalities.
Design partner program
We're opening POSTMAN to a small group of design partners: pharma, biotech, and research teams who want early access and a direct line to shape the product.
As a design partner, you get:
- Early access to POSTMAN before general availability
- Influence on the roadmap. Your use cases drive what we build next.
- Priority support from the Strand AI team, including custom evaluation on your data
- Co-publication opportunities on results and benchmarks
We're looking for teams with H&E slide archives who want to get spatial protein data from them, whether for biomarker discovery, patient stratification, clinical trial design, or translational research.
Apply for early access and we'll be in touch.