Stochastic modeling and neural network parameter inference for FRAP data analysis

smnnFRAP permits inference of kinetic parameters from FRAP data (diffusion coefficient, bound fraction and residence time). It contains two main modules:

• Stochastic model module: simulate a stochastic hybrid model of FRAP experiments to acquire in silico FRAP curves
• Parameter inference: using a library of simulated FRAP curves, perform parameter inference on FRAP data

To simulate the model simply run:

[trec,xrec,qrec,brec]=simulation_code(d,res_time,Fimm)

where: d the diffusion coefficient sigma, res_time the protein residence time and Fimm the protein bound fraction.

smnnFRAP was developed as a collaboration between the Cell Cycle Lab of the University of Patras, the IBIS team at INRIA-Grenoble – Rhône-Alpes and the Automatic Control Lab at ETH Zurich. It is freely distributed, available under the General Public License version 3 (GPL v3).

If you find our work useful please cite:

Rapsomaniki MA, Cinquemani E, Giakoumakis NN, Kotsantis P, Lygeros J, & Lygerou Z (2015). Inference of protein kinetics by stochastic modeling and simulation of fluorescence recovery after photobleaching experiments. Bioinformatics, 2015 Feb 1;31(3):355-62. (https://doi.org/10.1093/bioinformatics/btu619)