[gmx-users] For free energy landscape!

[dxli75]

Dear All,

 

    I have make a free energy landscape by using the data of RMSD and H-bond number of main chain of a protein according to the following method. I am not sure if the method does work. If not, Could you give me some advises? Thank you!

PS: Method

   1. Obtained the RMSD and the number of H-bond of mainchain of the protein by using g_rms and g_hbond commonds;

   2. Copy above two groups of values (RMSD and H-bond number) into Origin Software. first column is values of H-bond number, second column is RMSD;

3. Statistics > Descriptive Statistics > 2D Frequency Count in Origin, and obtained a matrix named as"TwoD FreqCounts";

4. Convert the above matrix into worksheet with type of xyz, and  got the values in the third column (frequency counts of RMSD and H-bond number);

5. Calculate the frequency count into normalized possibility;

6. Select the maximal or minimal possibility as control value.

   If  you select the maximal possibility (Pmax) as control value, the ΔG = Gx - Gmax = -kBTln(Px/Pmax) > 0;

   If  you select the minimal possibility (Pmin) as control value, the ΔG = Gx - Gmin = -kBTln(Px/Pmin) < 0;

Calculate all the values of (Px/Pmax) or (Px/Pmin) as the new third column. Here I choose Pmin (frequency count = 1) as control value, and get negative values of free energies;   

7. Convert the values in above three columns into a new matrix by using the method of XYZ Gridding. 

8. According to the above new matrix, plot the final free energy landscape.    

 


--

Dx Li
---------------------------------------
Mechanic Engineering Department
College of Engineering and Computing
University of South Carolina
300, Main Street, 29208
Columbia, SC
Tel: 803-9316307(M); 803-7771299(O)
E-mail:Daixi at cec.sc.edu
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20090819/1f4c924b/attachment.html>