NK Cell Missing-Self Recognition — Signal-Space Map
Interactive 2D signal-space simulation of natural killer cell decision-making: every target cell is plotted by its MHC-I (inhibitory) and stress-ligand (activating) level, a linear decision boundary separates the kill zone from the spare zone, and NK scanner agents traverse the map deciding each cell's fate.
Natural killer cells decide, target by target, whether to kill by weighing an inhibitory MHC-I signal against an activating stress-ligand signal. Rather than a rendered field of tissue, this view plots the whole target population directly in the two-dimensional space of those signals: MHC-I on one axis, stress ligand on the other. A straight decision-boundary line — the same rheostat rule the biology uses — separates the plane into a kill zone and a spare zone, and NK scanner agents sweep across this signal-space map, docking on the nearest cell and reading off its verdict. Downregulate MHC-I across the population, or carve out an infected patch, and watch points slide across the line: missing-self recognition made visible as pure geometry.
A 2D signal-space companion to the 3D NK cell scanner: every target cell is plotted by its MHC-I (inhibitory) and stress-ligand (activating) level, a linear decision boundary divides the plane into kill and spare zones, and NK scanner agents sweep the map itself, docking on the nearest cell by signal-space distance and reading off the missing-self verdict.
2D · HTML5 Canvas 2D · 60 FPS target · runs fully client-side, no install