Online help for 'Analysis of scanning force microscopy images of protein-induced DNA bending using simulations'


(Accompanying an article published in Nucleic Acids Research)


 
This method requires the measurement of end-to-end distance (EED) and contour length (L) of individual 1) bare DNA control molecules and 2) protein-DNA complexes. The protein must be bound at a specific position along the DNA contour.

Data files must be in text format with contour length data in column 1 and EED data in column 2. Two example data files are supplied along with the program.
Step-by-step instructions:
  1. Select the simulated data set that corresponds with the relative position of the protein-binding site along the DNA contour. (Every EED/L histogram is calculated over 100,000 simulated molecules.)
  2. Select the data files for the bare DNA control and the protein-DNA complex by pressing the 'folder' icon in the respective windows and choosing the desired file.
  3. Load the data files by pressing 'load and analyze files' or F5.
The graph 'L/P fit' on the 'data fitting' tab now shows the data of the bare DNA control set and an initial automatic fit of these data to the simulated DNA data set (with the accuracy of the fit depicted in the χ2 graph below). The graph 'angle fit' analogously shows the data of the protein-DNA complexes fit to the simulated DNA data set (and in the χ2 graph below the associated error).

The binning of the data sets may be optimized manually by changing the number of bins. If a mixed population of bare DNA molecules and protein-DNA complexes is used, the uncomplexed fraction can be manually adapted.
  1. The fitting parameters (P and the DNA bending angle) for the used data sets can now be read directly from their respective windows.
If needed, the data within the graphs can be exported to a text file for further processing.


To directly visualize the simulation results, one can select the 'display simulations data set' tab. By changing the value of either the L/P value or the angle controls, a section of the simulation results is shown in the two graphs.


For any further questions, please e-mail Remus Th. Dame or Joost van Mameren.


If the use of this analysis method leads to publication of your results, please refer to Dame, R.T., van Mameren. J. et al., Nucleic Acids Res. 33: e68.