Huygens FRAP Simulator
Table of contents
1 Introduction
The Huygens FRAP Simulator (Figure 1) is an external module for the Huygens software. It simulates the Fluorescence Recovery After Photobleaching (FRAP) method and can compare it with FRAP curves from experiments [1].FRAP is an optical technique used to determine the diffusion speed of certain particles. In this method all particles of the desired type are labelled with a photosensitive part and a specific area of the specimen is bleached with a high intensity laser. This disables the fluorophores permanently. The intensity of the bleached area is then measured over time as particles with an active dye diffuse into the bleached area.
When the areas is scanned with a low intensity beam, this
recovery can be measured over time.
The steepness of the resulting intensity recovery curve is related to the mobility of the particles.
In order to translate this curve to actual diffusion parameters, one has to define a diffusion model and associated parameters.
Firstly there is the amount of active particles left in the cell, which should be sufficiently numerous to have a statistically reliable diffusion. Secondly, there is the beam profile and thus the 3D shape of the bleached spot, which is not a sphere or box when viewed, but more or less an hourglass shape.
Lastly, the size and shape of the cell determine the actual diffusion. Particles of interest can get stuck on their location for a while, either due to obstructions or because of bonding to a specific site. If a significant fraction of the particles is stuck around the time of bleaching, this of course affects the curve, suggesting a lower diffusion than is actually the case. Using this simulator it is possible to estimate both sticking parameters and diffusion from a measured curve.
Figure 1: The FRAP simulator.
2 Method
A Monte Carlo method is used to simulate the Fluorescence Recovery After Photobleaching. The random numbers used by the simulation are generated with the Mersenne Twister [2] pseudorandom number generator (PRNG). Random movements of the particles are computed using the Ziggurat method [3]. To estimate both sticking and diffusion parameters from a measured curve the simulation fits the curve using the NelderMead method [4].3 Installation
The FRAP Simulator can be downloaded here.Requirements for installing the module are Linux, gcc, OpenMP and Huygens. Installation is handled by the Makefile in the root dir. Running make compiles the module and copies it to your SVI Plugins directory.
4 Usage
The module should be placed in the Plugins directory in your SVI directory. To load the module Huygens should be started with the argument -checkLoadMod. After the module is loaded you can start the simulator using the following command in the Huygens Tcl shell:::FrapSimulator::create
4.1 Set parameters
The FRAP simulator has the following parameters:4.1.1 Simulation parameters
Simulation mode
Parameter to set the simulation mode. Simulate runs a simulation using the specified parameters. Simulate & compare runs a simulation using the specified parameters and compares it to existing curve. It requires a file containing an existing curve to compare to. Autofit will try to fit the simulated curve by changing the diffusion, stick probability and stick time. This may take long and requires a compare file with an existing curve.
Number of particles
The number of particles used in the simulation. More particles make it more accurate, but also slower. Advised: 100k to 10 million particles.
Length of simulation
Total amount of iterations in the whole simulation (including pre-bleaching period). On every iteration all particles are moved once.
Number of threads
Specifies the number of threads used for the simulation. Advised: as many threads as you have CPU cores available or a multiple of this.
Compare file
A tab separated file with on every row a single data point. The time in seconds from the start of the experiment and the normalized intensity of the radiated area. Data is in chronological order.
Output directory
Directory to store resulting curve and particles count.
Debug messages
Show debug messages if Huygens is started from the command line.
4.1.2 Physical parameters
Diffusion
Average diffusion of the particles per iteration. Advised: 0.5 to 2.0.
Bleach delay
Number of iterations the simulation should run to reach equilibrium before bleaching. Advised: 100 to 1000 iterations (220 iterations = 5 seconds).
Bleach probability
Probability a particle gets bleached if it is in the radiated area. Advised: 1.0 to 3.0.
Stick probability
The probability per iteration a particle gets stuck at its current location. Advised: 0.0 to 0.1.
Stick time
The average number of iterations a particle remains stuck on its location if it gets stuck. Note that it can still be blinked during this time. Advised: 0 to 100.
Blink probability
The probability per iteration a particle blinks (becomes invisible and untouchable for bleaching). Advised: 0.0 to 0.1 (0.0007 for GFP).
Blink time
The average number of iterations a particle remains in blinked state once it gets blinked. Note that it can still get stuck during this time. Advised: 0 to 200 (75 for GFP).
4.2 Run simulation
After setting the parameters the simulation can be started by clicking the start simulation button. If there are parameters that are incomplete or incorrect, the simulation will not be started until they are corrected. Labels of incomplete or incorrect parameters are shown in red. When the simulation is finished the simulation curve and an averaged version of the curve is plotted in the data plotter. If the simulations run in Autofit or Simulate & Compare mode also the curve from the compare file is plotted. The resulting parameters of an autofit simulation are set in the parameter fields.5 Acknowledgment
SVI wishes to thank Sebastian Oude Voshaar for his work on the simulator during his intership.Thanks to Adriaan Houtsmuller and Gert van Capellen from Erasmus MC for the data and for useful discussions about the simulator.
6 License
This program is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License (as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.This program is distributed in the hope that it will be useful, but without any warranty ; without even the implied warranty of merchantability or fitness for a particular purpose. See the GNU General Public License for more details.
You should have received a copy of the GNU General Public License along with this program. If not, see www.gnu.org/licenses/.
References
1 Martin E. van Royen, Pascal Farla, Karin A. Mattern, Bart Geverts, Jan Trapman, Adriaan B. Houtsmuller. (2008). Fluorescence Recovery After Photobleaching (FRAP) to Study Nuclear Protein Dynamics in Living Cells, Humana Press, 464: 363–385.2 M. Matsumoto en T. Nishimura. (1998). Mersenne twister: A 623-dimensionally equidistributed uniform pseudorandom number generator, ACM Trans. on Modeling and Computer Simulations.
3 G. Marsaglia and W.W. Tsang. (2000). The Ziggurat Method for Generating Random Variables, Journal of Statistical Software, 5 (8): 1–7.
4 J. A. Nelder and R. Mead. (1965). A Simplex Method for Function Minimization, The Computer Journal, 7 (4): 308–313.