3D UFO Cell Culture Dose-Response Simulator

Claimed Exposure

Claimed dose 1.0 Gy

Cell Radiosensitivity (LQ model)

α — linear term 0.20 /Gy

β — quadratic term 0.020 /Gy²

Assay Design

Background mutation rate 0.20%

Statistical sample size 300

The 16×16 dish grids are a visual sample (256 cells each); the z-test below uses the full statistical sample size.

Predicted survival S(D)

–

Dish survival (live tally)

–

Exposed dish dead

–

Control dish dead

–

z-score (n = full sample)

–

Verdict

–

How it works

The same linear-quadratic dose-response math as the 2D chart version drives this scene, but staged as two physical cell cultures instead of a curve: an exposed dish, dosed at the claimed exposure, and an unexposed control dish. Each of the 256 cells in a dish is an independent coin-flip against the model's predicted damage probability — watch individual cells actually die (shrink and redden) rather than reading a single aggregate percentage.

S(D) = e^(−αD − βD²)
per-cell: dies with probability p = 1 − S(D)  (exposed)
                              p = background rate  (control)
z = (p̂₁ − p̂₂) / sqrt(p̂·(1−p̂)·(1/n₁ + 1/n₂))

Exposed dish (left) — each cell independently dies with the LQ model's predicted probability at the claimed dose; raise dose or radiosensitivity and watch more spheres shrink and turn red.
Control dish (right) — cells die only at the background mutation rate, with zero radiation dose — this is the null-hypothesis baseline the exposed dish is judged against.
Irradiate dishes — re-rolls both dishes from the current probabilities, and also redraws the full-sample z-test below so you can see how a small visual sample can look dramatic while the statistically rigorous test on the full assay says otherwise (or vice versa).
Dish survival tally — the live fraction of the 256-cell visual sample still alive, shown alongside the model's exact predicted S(D) so you can see sampling noise directly on a finite dish.

This is the visual, cell-by-cell face of the same statistical test the 2D version runs numerically: real evidence of radiation-induced cellular damage requires the observed rate to clear the background noise floor by a statistically significant margin, not just look worse in a small dish.

Living cell Dead / damaged cell Dish (culture plate)

⚠ Couldn't load the 3D engineThree.js failed to load from the CDN. Check your connection and reload.