What problem does it solve?
Manually inspecting marker columns spot by spot to assign tissue regions is slow and error-prone. This Skill deterministically maps each spot in a local CSV to a region label (immune_edge, tumor_core, stromal_zone, proliferative_core) based on the dominant marker among EPCAM, PTPRC, COL1A1, and MKI67, then produces a report, tables, and an SVG map.
Core Features & Use Cases
- Deterministic region assignment: Maps each spot to a region using the highest of four marker counts, with no external APIs or uploads.
- Hotspot flagging: Flags tumor-core and MKI67-dominant proliferative-core spots for review.
- Structured outputs: Writes report.md, result.json, mapped_spots.csv, region_summary.csv, an SVG spot map, and a reproducibility command log.
- Use Case: A researcher with a spot-level marker count table runs the demo or their own CSV to get a labeled region map and summary tables for downstream figure integration.
Quick Start
Run the marker dominance mapper on your spot-count CSV by asking the agent to map marker-dominance spots from your file, or use the built-in demo mode to see a synthetic six-spot region map.