loading...
running...
How to use
Pick a preset or set D, the mobile fraction, the spot size and the cell. The pretend confocal microscope records a pre-bleach image, fires the bleach pulse and then watches the spot refill. The recovery curve compares three independent calculations: the closed form (green), the finite-volume PDE (orange) and random-walking molecules (blue dots). The profile graph shows the brightness across the spot. Use Measure once to fit a noisy virtual data set and see how well D and Mf are recovered, and Repeat 40 times for the scatter. Space pauses, R restarts. Educational model, not for lab analysis.
The 3D page shows the microscope, a heavily magnified adherent cell on a coverslip and a monitor with the live recovery curve. The glowing patch on the cell is the PDE solution painted on the dome; the dots are individual random-walking molecules (bright = unbleached, dark red = bleached). The microscope stand, objective, stage, cell, laser box and monitor are Blender 5.2 models with baked PBR textures. Drag to orbit, wheel to zoom.
The science
FRAP (fluorescence recovery after photobleaching) measures how fast fluorescent molecules move. A strong laser pulse destroys the dye in a small spot of a cell membrane; unbleached molecules from outside diffuse in, and the spot brightens again. The speed of that recovery gives the diffusion coefficient D, and the level it reaches gives the mobile fraction.
Model. The membrane is a flat disc of radius R. The mobile pool (fraction Mf) obeys 2D diffusion; u(r,t) is its unbleached fraction:
du/dt = D (d2u/dr2 + (1/r) du/dr), u(r,0) = 1 - K for r < w, 1 outside, du/dr = 0 at r = R
w is the spot radius and K the bleach depth. The spot intensity is I = Mf <u>spot + (1 - Mf)(1 - K) (the immobile molecules inside the spot stay dark) and the normalised recovery is
F(t) = (I(t) - I(0)) / (I(-) - I(0)) = Mf (<u>spot - (1 - K)) / K
Closed form. For a uniform circular spot in an infinite membrane (Soumpasis, 1983):
F(t) = Mf exp(-2 tau/t) [ I0(2 tau/t) + I1(2 tau/t) ], tau = w^2 / (4 D)
t_1/2 = 0.2236 w^2 / D => D = 0.2236 w^2 / t_1/2
The recovery time grows as w^2: a spot half as wide recovers four times faster. At the spot centre the exact solution is u(0,t) = 1 - K (1 - exp(-w^2 / 4Dt)). In a closed disc the final level is F(inf) = Mf (1 - (w/R)^2), because the bleached molecules never leave the cell.
Read-out bleaching. Every frame also bleaches the dye a little. Recorded F is multiplied by exp(-k t); dividing by a reference region (or the whole-cell signal) removes it. Without that correction the plateau, and so the mobile fraction, comes out too low.
Numerics. The PDE is solved by an exactly conservative finite-volume scheme (Crank-Nicolson after a few implicit start-up steps). The blue dots come from an independent method: thousands of random-walking molecules with Gaussian steps of variance 2 D dt (Einstein relation <r^2> = 4 D t).
Honest simplifications. Flat 2D membrane (a real cell has a dome, ruffles and a nucleus; the 3D page draws the dome but the maths ignores its curvature); pure diffusion with no binding or reaction (real FRAP often shows binding, which needs more parameters); a sharp-edged uniform bleach disc (real beams are closer to Gaussian and bleach during the pulse, so published constants differ: for a Gaussian beam of 1/e^2 radius w, D is about 0.88 w^2 / (4 t_1/2), and the constant depends on the bleach depth); immobile molecules are perfectly static; no dye blinking, no flow or active transport; the D range is chosen for teaching, not for a specific protein. Typical literature values, quoted from memory: lipid analogues in cell membranes 0.5 to 2 um^2/s, transmembrane proteins 0.01 to 0.5 um^2/s, GFP in cytosol 20 to 30 um^2/s (a 3D problem, shown here only as a fast 2D idealisation).