Single-cell & spatial · R&D demonstration, public data · May 2026
Spatial Mapping of the Tumor Microenvironment
Localizing immune and stromal cell populations in tumor tissue using spatial deconvolution on a public 10x Visium dataset.
Challenge
Localizing immune and stromal cell populations within tumor tissue architecture, in a way that respects spatial context rather than treating cells as a loose bag of transcripts.
Approach
Spatial deconvolution and neighborhood analysis on a public Visium tumor dataset, integrated with a matched scRNA-seq reference.
What we found
The analysis resolved several spatially distinct niches within the tumor margin. The clustering resolution and validation approach behind that count aren't published alongside the figure yet — see the caveat on the metric above.
Limits of this demonstration
This is an R&D demonstration on a public 10x Genomics Visium dataset, not a client
engagement. No repo or notebook is linked yet, which is why the metric above is flagged
TODO-VERIFY. Visium's spot resolution mixes multiple cells per spot, which limits how
finely any niche boundary here can actually be trusted.