A Bayesian method for inferring quantitative information from FRET data

Matlab files: FRETcode.zip

(Right click or left click and use the `File' menu to download. Octave is a free equivalent to Matlab.)


  1. Generate sample data by typing the following in Matlab:

    >> [datasim,cparams]= datamaker;

    Three bar plots of the sample data (one for each cell) will be displayed:

    This sample data comes from 3 cells with 10 measurements per cell and 5% measurement noise.

  2. Run the analysis, generating the MCMC walk and the contour plot of the energy, by typing:

    >> runMCMCandContour

    The code will take a few minutes to run. When finished, it will display various plots of the results, for instance:

    In this example, the true values are Kd= 1e-6 M and Efr= 0.4.

Other options:

  1. To experiment with other data (e.g. with different numbers of cells, numbers of measurements, or levels of measurement noise), include arguments when calling datamaker:

    >> [data,cparams]= datamaker(Efr,Kd,n,A0mu,D0mu,r,sig,plot_on);

    Further details about these arguments can be found by typing

    >> help datamaker

  2. To adjust parameters of the analysis (e.g. stepsize and number of steps for the MCMC, its starting location, prior information, and number of parameters to fit) edit section 1 in the script runMCMCandContour.m.